Hugging Face's logo

FidelityWM
/
planning-baselines

world-model
planning
dinowm
pldm
fast-lewm
Model card Files
xet
 Community
greycat111 commited on
Commit
8e66f17
·
verified ·
1 Parent(s): 4aaa3c7

Complete PLDM Cube five-seed planning evaluation and report mean/std

Browse files
Files changed (30) hide show
  1. README.md +1 -1
  2. pldm/cube/README.md +2 -4
  3. pldm/cube/provenance.json +15 -1
  4. results/completed_seed_results.csv +4 -0
  5. results/completed_seed_results.md +5 -1
  6. results/planning_success.csv +80 -80
  7. results/planning_success.json +220 -220
  8. results/planning_success.xlsx +0 -0
  9. results/planning_summary.csv +4 -4
  10. results/planning_summary.json +36 -36
  11. results/pldm_cube_100.json +312 -0
  12. results/pldm_cube_100_seed43.json +312 -0
  13. results/pldm_cube_100_seed44.json +312 -0
  14. results/pldm_cube_100_seed45.json +312 -0
  15. results/pldm_cube_100_seed46.json +312 -0
  16. results/pldm_cube_25.json +312 -0
  17. results/pldm_cube_25_seed43.json +312 -0
  18. results/pldm_cube_25_seed44.json +312 -0
  19. results/pldm_cube_25_seed45.json +312 -0
  20. results/pldm_cube_25_seed46.json +312 -0
  21. results/pldm_cube_50.json +312 -0
  22. results/pldm_cube_50_seed43.json +312 -0
  23. results/pldm_cube_50_seed44.json +312 -0
  24. results/pldm_cube_50_seed45.json +312 -0
  25. results/pldm_cube_50_seed46.json +312 -0
  26. results/pldm_cube_75.json +312 -0
  27. results/pldm_cube_75_seed43.json +312 -0
  28. results/pldm_cube_75_seed44.json +312 -0
  29. results/pldm_cube_75_seed45.json +312 -0
  30. results/pldm_cube_75_seed46.json +312 -0
README.md CHANGED
@@ -14,7 +14,7 @@ This repository archives checkpoints and measured planning results for TwoRoom,
14
  |---|---|---|
15
  | fast-lewm/tworoom, cube, pusht | Exact original `naiverer/fast-leworldmodel` checkpoints | All five-seed planning evaluations completed |
16
  | dinowm/pusht | Original DINO-WM OSF PushT checkpoint and frozen DINOv2-S/14 backbone | Published model trained on `pusht_noise`; evaluation incomplete |
17
- | pldm/cube | Final epoch-10 model, epoch-8 resume checkpoint, configuration and provenance | Training complete; planning evaluation not run |
18
  | pldm/tworoom | Configuration and results only | The trained weights were deleted after evaluation under the earlier disk-saving instruction and are unavailable |
19
 
20
  Newly completed local checkpoints are archived under `<method>/<task>/model.pt`, with provenance alongside each file. Each local checkpoint uses training seed 42. Five evaluation seeds do not mean five independently trained models.
 
14
  |---|---|---|
15
  | fast-lewm/tworoom, cube, pusht | Exact original `naiverer/fast-leworldmodel` checkpoints | All five-seed planning evaluations completed |
16
  | dinowm/pusht | Original DINO-WM OSF PushT checkpoint and frozen DINOv2-S/14 backbone | Published model trained on `pusht_noise`; evaluation incomplete |
17
+ | pldm/cube | Final epoch-10 model, epoch-8 resume checkpoint, configuration and provenance | Training and all five-seed planning evaluations complete |
18
  | pldm/tworoom | Configuration and results only | The trained weights were deleted after evaluation under the earlier disk-saving instruction and are unavailable |
19
 
20
  Newly completed local checkpoints are archived under `<method>/<task>/model.pt`, with provenance alongside each file. Each local checkpoint uses training seed 42. Five evaluation seeds do not mean five independently trained models.
pldm/cube/README.md CHANGED
@@ -1,7 +1,5 @@
1
  # PLDM Cube checkpoints
2
 
3
- - `model.pt`: final 10-epoch model (training seed 42); recommended for loading the trained model. SHA256 and training details are in `provenance.json`.
4
- - `resume_checkpoint.pt`: earlier epoch-8 training snapshot including optimizer and RNG state, approximately 217 MB. Its provenance is in `resume_checkpoint.provenance.json`.
5
- - `config.json`: training configuration.
6
 
7
- Trained on the supplied `cube_single_expert.h5`. Planning evaluation of the final model was not started: training finished, the final weights were uploaded, and the program stopped at the user's request.
 
1
  # PLDM Cube checkpoints
2
 
3
+ `model.pt` is the final 10-epoch model trained with seed 42. All five evaluation seeds (42–46) and four goal offsets (25, 50, 75, 100) now have 200 completed episodes each. See ../../results/planning_success.xlsx for per-seed scores and mean ± sample standard deviation.
 
 
4
 
5
+ `resume_checkpoint.pt` is the older epoch-8 snapshot with optimizer and RNG state. Its metadata is in `resume_checkpoint.provenance.json`. Final-model metadata is in `provenance.json`.
pldm/cube/provenance.json CHANGED
@@ -3,5 +3,19 @@
3
  "epochs_completed": 10,
4
  "training_seed": 42,
5
  "variant": "PLDM stable-worldmodel recipe",
6
- "status": "Final 10-epoch checkpoint; planning evaluation not run; stopped at user request"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  }
 
3
  "epochs_completed": 10,
4
  "training_seed": 42,
5
  "variant": "PLDM stable-worldmodel recipe",
6
+ "status": "Final epoch-10 model; all 20 planning runs completed",
7
+ "evaluation_seeds": [
8
+ 42,
9
+ 43,
10
+ 44,
11
+ 45,
12
+ 46
13
+ ],
14
+ "goal_offsets": [
15
+ 25,
16
+ 50,
17
+ 75,
18
+ 100
19
+ ],
20
+ "samples_per_run": 200
21
  }
results/completed_seed_results.csv CHANGED
@@ -3,6 +3,10 @@ pldm,tworoom,25,96.0,94.5,95.5,94.0,97.0,5,95.4,1.1937336386313322,95.40 ± 1.19
3
  pldm,tworoom,50,80.0,76.0,81.0,78.5,79.5,5,79.0,1.9039432764659772,79.00 ± 1.90
4
  pldm,tworoom,75,57.5,54.5,52.5,54.5,56.0,5,55.0,1.8708286933869707,55.00 ± 1.87
5
  pldm,tworoom,100,33.5,33.5,33.5,34.5,32.0,5,33.4,0.8944271909999159,33.40 ± 0.89
 
 
 
 
6
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
7
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
8
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
 
3
  pldm,tworoom,50,80.0,76.0,81.0,78.5,79.5,5,79.0,1.9039432764659772,79.00 ± 1.90
4
  pldm,tworoom,75,57.5,54.5,52.5,54.5,56.0,5,55.0,1.8708286933869707,55.00 ± 1.87
5
  pldm,tworoom,100,33.5,33.5,33.5,34.5,32.0,5,33.4,0.8944271909999159,33.40 ± 0.89
6
+ pldm,cube,25,55.0,52.0,56.0,56.5,66.0,5,57.1,5.272570530585627,57.10 ± 5.27
7
+ pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
8
+ pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
9
+ pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
10
  fast-lewm,tworoom,25,88.0,90.0,86.5,88.5,91.0,5,88.8,1.753567791675018,88.80 ± 1.75
11
  fast-lewm,tworoom,50,64.0,58.0,59.0,63.5,66.0,5,62.1,3.4351128074635335,62.10 ± 3.44
12
  fast-lewm,tworoom,75,44.5,33.5,35.5,34.0,38.0,5,37.1,4.491658936295141,37.10 ± 4.49
results/completed_seed_results.md CHANGED
@@ -1,6 +1,6 @@
1
  # Completed five-seed planning results
2
 
3
- Success percentages; 200 episodes per seed and offset. Sample standard deviation (ddof=1). These are evaluation seeds of one checkpoint per model/task, not independently trained model seeds.
4
 
5
  | Model | Dataset | Offset | Seed 42 | Seed 43 | Seed 44 | Seed 45 | Seed 46 | Mean ± std |
6
  |---|---|---:|---:|---:|---:|---:|---:|---:|
@@ -8,6 +8,10 @@ Success percentages; 200 episodes per seed and offset. Sample standard deviation
8
  | pldm | tworoom | 50 | 80.0 | 76.0 | 81.0 | 78.5 | 79.5 | 79.00 ± 1.90 |
9
  | pldm | tworoom | 75 | 57.5 | 54.5 | 52.5 | 54.5 | 56.0 | 55.00 ± 1.87 |
10
  | pldm | tworoom | 100 | 33.5 | 33.5 | 33.5 | 34.5 | 32.0 | 33.40 ± 0.89 |
 
 
 
 
11
  | fast-lewm | tworoom | 25 | 88.0 | 90.0 | 86.5 | 88.5 | 91.0 | 88.80 ± 1.75 |
12
  | fast-lewm | tworoom | 50 | 64.0 | 58.0 | 59.0 | 63.5 | 66.0 | 62.10 ± 3.44 |
13
  | fast-lewm | tworoom | 75 | 44.5 | 33.5 | 35.5 | 34.0 | 38.0 | 37.10 ± 4.49 |
 
1
  # Completed five-seed planning results
2
 
3
+ Success percentages, 200 episodes per seed and offset. Sample standard deviation (ddof=1). Seeds are evaluation seeds, not independent model training runs.
4
 
5
  | Model | Dataset | Offset | Seed 42 | Seed 43 | Seed 44 | Seed 45 | Seed 46 | Mean ± std |
6
  |---|---|---:|---:|---:|---:|---:|---:|---:|
 
8
  | pldm | tworoom | 50 | 80.0 | 76.0 | 81.0 | 78.5 | 79.5 | 79.00 ± 1.90 |
9
  | pldm | tworoom | 75 | 57.5 | 54.5 | 52.5 | 54.5 | 56.0 | 55.00 ± 1.87 |
10
  | pldm | tworoom | 100 | 33.5 | 33.5 | 33.5 | 34.5 | 32.0 | 33.40 ± 0.89 |
11
+ | pldm | cube | 25 | 55.0 | 52.0 | 56.0 | 56.5 | 66.0 | 57.10 ± 5.27 |
12
+ | pldm | cube | 50 | 49.5 | 39.0 | 41.5 | 44.5 | 50.5 | 45.00 ± 4.97 |
13
+ | pldm | cube | 75 | 54.5 | 58.5 | 57.0 | 54.5 | 51.5 | 55.20 ± 2.68 |
14
+ | pldm | cube | 100 | 57.5 | 58.0 | 51.5 | 54.0 | 53.0 | 54.80 ± 2.84 |
15
  | fast-lewm | tworoom | 25 | 88.0 | 90.0 | 86.5 | 88.5 | 91.0 | 88.80 ± 1.75 |
16
  | fast-lewm | tworoom | 50 | 64.0 | 58.0 | 59.0 | 63.5 | 66.0 | 62.10 ± 3.44 |
17
  | fast-lewm | tworoom | 75 | 44.5 | 33.5 | 35.5 | 34.0 | 38.0 | 37.10 ± 4.49 |
results/planning_success.csv CHANGED
@@ -1,44 +1,44 @@
1
  model,dataset,offset,seed,samples_completed,samples_required,successes,success_percent,status,execution_budget,checkpoint_release,training_epochs,training_seed,model_variant
2
- dinowm,tworoom,25,42,0,200,,,queued training,50,,,,
3
- dinowm,tworoom,25,43,0,200,,,queued training,50,,,,
4
- dinowm,tworoom,25,44,0,200,,,queued training,50,,,,
5
- dinowm,tworoom,25,45,0,200,,,queued training,50,,,,
6
- dinowm,tworoom,25,46,0,200,,,queued training,50,,,,
7
- dinowm,tworoom,50,42,0,200,,,queued training,50,,,,
8
- dinowm,tworoom,50,43,0,200,,,queued training,50,,,,
9
- dinowm,tworoom,50,44,0,200,,,queued training,50,,,,
10
- dinowm,tworoom,50,45,0,200,,,queued training,50,,,,
11
- dinowm,tworoom,50,46,0,200,,,queued training,50,,,,
12
- dinowm,tworoom,75,42,0,200,,,queued training,50,,,,
13
- dinowm,tworoom,75,43,0,200,,,queued training,50,,,,
14
- dinowm,tworoom,75,44,0,200,,,queued training,50,,,,
15
- dinowm,tworoom,75,45,0,200,,,queued training,50,,,,
16
- dinowm,tworoom,75,46,0,200,,,queued training,50,,,,
17
- dinowm,tworoom,100,42,0,200,,,queued training,50,,,,
18
- dinowm,tworoom,100,43,0,200,,,queued training,50,,,,
19
- dinowm,tworoom,100,44,0,200,,,queued training,50,,,,
20
- dinowm,tworoom,100,45,0,200,,,queued training,50,,,,
21
- dinowm,tworoom,100,46,0,200,,,queued training,50,,,,
22
- dinowm,cube,25,42,0,200,,,queued training,50,,,,
23
- dinowm,cube,25,43,0,200,,,queued training,50,,,,
24
- dinowm,cube,25,44,0,200,,,queued training,50,,,,
25
- dinowm,cube,25,45,0,200,,,queued training,50,,,,
26
- dinowm,cube,25,46,0,200,,,queued training,50,,,,
27
- dinowm,cube,50,42,0,200,,,queued training,50,,,,
28
- dinowm,cube,50,43,0,200,,,queued training,50,,,,
29
- dinowm,cube,50,44,0,200,,,queued training,50,,,,
30
- dinowm,cube,50,45,0,200,,,queued training,50,,,,
31
- dinowm,cube,50,46,0,200,,,queued training,50,,,,
32
- dinowm,cube,75,42,0,200,,,queued training,50,,,,
33
- dinowm,cube,75,43,0,200,,,queued training,50,,,,
34
- dinowm,cube,75,44,0,200,,,queued training,50,,,,
35
- dinowm,cube,75,45,0,200,,,queued training,50,,,,
36
- dinowm,cube,75,46,0,200,,,queued training,50,,,,
37
- dinowm,cube,100,42,0,200,,,queued training,50,,,,
38
- dinowm,cube,100,43,0,200,,,queued training,50,,,,
39
- dinowm,cube,100,44,0,200,,,queued training,50,,,,
40
- dinowm,cube,100,45,0,200,,,queued training,50,,,,
41
- dinowm,cube,100,46,0,200,,,queued training,50,,,,
42
  dinowm,pusht,25,42,24,200,21,,running,50,original DINO-WM OSF outputs/pusht; trained on pusht_noise,,,
43
  dinowm,pusht,25,43,0,200,,,queued,50,,,,
44
  dinowm,pusht,25,44,0,200,,,queued,50,,,,
@@ -79,46 +79,46 @@ pldm,tworoom,100,43,200,200,67,33.5,complete,50,Locally trained 10 epochs: PLDM
79
  pldm,tworoom,100,44,200,200,67,33.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
80
  pldm,tworoom,100,45,200,200,69,34.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
81
  pldm,tworoom,100,46,200,200,64,32.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
82
- pldm,cube,25,42,0,200,,,training,50,,,,
83
- pldm,cube,25,43,0,200,,,training,50,,,,
84
- pldm,cube,25,44,0,200,,,training,50,,,,
85
- pldm,cube,25,45,0,200,,,training,50,,,,
86
- pldm,cube,25,46,0,200,,,training,50,,,,
87
- pldm,cube,50,42,0,200,,,training,50,,,,
88
- pldm,cube,50,43,0,200,,,training,50,,,,
89
- pldm,cube,50,44,0,200,,,training,50,,,,
90
- pldm,cube,50,45,0,200,,,training,50,,,,
91
- pldm,cube,50,46,0,200,,,training,50,,,,
92
- pldm,cube,75,42,0,200,,,training,50,,,,
93
- pldm,cube,75,43,0,200,,,training,50,,,,
94
- pldm,cube,75,44,0,200,,,training,50,,,,
95
- pldm,cube,75,45,0,200,,,training,50,,,,
96
- pldm,cube,75,46,0,200,,,training,50,,,,
97
- pldm,cube,100,42,0,200,,,training,50,,,,
98
- pldm,cube,100,43,0,200,,,training,50,,,,
99
- pldm,cube,100,44,0,200,,,training,50,,,,
100
- pldm,cube,100,45,0,200,,,training,50,,,,
101
- pldm,cube,100,46,0,200,,,training,50,,,,
102
- pldm,pusht,25,42,0,200,,,queued training,50,,,,
103
- pldm,pusht,25,43,0,200,,,queued training,50,,,,
104
- pldm,pusht,25,44,0,200,,,queued training,50,,,,
105
- pldm,pusht,25,45,0,200,,,queued training,50,,,,
106
- pldm,pusht,25,46,0,200,,,queued training,50,,,,
107
- pldm,pusht,50,42,0,200,,,queued training,50,,,,
108
- pldm,pusht,50,43,0,200,,,queued training,50,,,,
109
- pldm,pusht,50,44,0,200,,,queued training,50,,,,
110
- pldm,pusht,50,45,0,200,,,queued training,50,,,,
111
- pldm,pusht,50,46,0,200,,,queued training,50,,,,
112
- pldm,pusht,75,42,0,200,,,queued training,50,,,,
113
- pldm,pusht,75,43,0,200,,,queued training,50,,,,
114
- pldm,pusht,75,44,0,200,,,queued training,50,,,,
115
- pldm,pusht,75,45,0,200,,,queued training,50,,,,
116
- pldm,pusht,75,46,0,200,,,queued training,50,,,,
117
- pldm,pusht,100,42,0,200,,,queued training,50,,,,
118
- pldm,pusht,100,43,0,200,,,queued training,50,,,,
119
- pldm,pusht,100,44,0,200,,,queued training,50,,,,
120
- pldm,pusht,100,45,0,200,,,queued training,50,,,,
121
- pldm,pusht,100,46,0,200,,,queued training,50,,,,
122
  fast-lewm,tworoom,25,42,200,200,176,88.0,complete,50,naiverer/fast-leworldmodel,,,
123
  fast-lewm,tworoom,25,43,200,200,180,90.0,complete,50,naiverer/fast-leworldmodel,,,
124
  fast-lewm,tworoom,25,44,200,200,173,86.5,complete,50,naiverer/fast-leworldmodel,,,
 
1
  model,dataset,offset,seed,samples_completed,samples_required,successes,success_percent,status,execution_budget,checkpoint_release,training_epochs,training_seed,model_variant
2
+ dinowm,tworoom,25,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
3
+ dinowm,tworoom,25,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
4
+ dinowm,tworoom,25,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
5
+ dinowm,tworoom,25,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
6
+ dinowm,tworoom,25,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
7
+ dinowm,tworoom,50,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
8
+ dinowm,tworoom,50,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
9
+ dinowm,tworoom,50,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
10
+ dinowm,tworoom,50,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
11
+ dinowm,tworoom,50,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
12
+ dinowm,tworoom,75,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
13
+ dinowm,tworoom,75,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
14
+ dinowm,tworoom,75,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
15
+ dinowm,tworoom,75,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
16
+ dinowm,tworoom,75,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
17
+ dinowm,tworoom,100,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
18
+ dinowm,tworoom,100,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
19
+ dinowm,tworoom,100,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
20
+ dinowm,tworoom,100,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
21
+ dinowm,tworoom,100,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
22
+ dinowm,cube,25,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
23
+ dinowm,cube,25,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
24
+ dinowm,cube,25,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
25
+ dinowm,cube,25,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
26
+ dinowm,cube,25,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
27
+ dinowm,cube,50,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
28
+ dinowm,cube,50,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
29
+ dinowm,cube,50,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
30
+ dinowm,cube,50,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
31
+ dinowm,cube,50,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
32
+ dinowm,cube,75,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
33
+ dinowm,cube,75,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
34
+ dinowm,cube,75,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
35
+ dinowm,cube,75,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
36
+ dinowm,cube,75,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
37
+ dinowm,cube,100,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
38
+ dinowm,cube,100,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
39
+ dinowm,cube,100,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
40
+ dinowm,cube,100,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
41
+ dinowm,cube,100,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
42
  dinowm,pusht,25,42,24,200,21,,running,50,original DINO-WM OSF outputs/pusht; trained on pusht_noise,,,
43
  dinowm,pusht,25,43,0,200,,,queued,50,,,,
44
  dinowm,pusht,25,44,0,200,,,queued,50,,,,
 
79
  pldm,tworoom,100,44,200,200,67,33.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
80
  pldm,tworoom,100,45,200,200,69,34.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
81
  pldm,tworoom,100,46,200,200,64,32.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
82
+ pldm,cube,25,42,200,200,110,55.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
83
+ pldm,cube,25,43,200,200,104,52.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
84
+ pldm,cube,25,44,200,200,112,56.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
85
+ pldm,cube,25,45,200,200,113,56.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
86
+ pldm,cube,25,46,200,200,132,66.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
87
+ pldm,cube,50,42,200,200,99,49.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
88
+ pldm,cube,50,43,200,200,78,39.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
89
+ pldm,cube,50,44,200,200,83,41.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
90
+ pldm,cube,50,45,200,200,89,44.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
91
+ pldm,cube,50,46,200,200,101,50.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
92
+ pldm,cube,75,42,200,200,109,54.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
93
+ pldm,cube,75,43,200,200,117,58.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
94
+ pldm,cube,75,44,200,200,114,57.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
95
+ pldm,cube,75,45,200,200,109,54.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
96
+ pldm,cube,75,46,200,200,103,51.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
97
+ pldm,cube,100,42,200,200,115,57.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
98
+ pldm,cube,100,43,200,200,116,58.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
99
+ pldm,cube,100,44,200,200,103,51.5,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
100
+ pldm,cube,100,45,200,200,108,54.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
101
+ pldm,cube,100,46,200,200,106,53.0,complete,50,Locally trained 10 epochs: PLDM stable-worldmodel recipe,10,42,PLDM stable-worldmodel recipe
102
+ pldm,pusht,25,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
103
+ pldm,pusht,25,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
104
+ pldm,pusht,25,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
105
+ pldm,pusht,25,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
106
+ pldm,pusht,25,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
107
+ pldm,pusht,50,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
108
+ pldm,pusht,50,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
109
+ pldm,pusht,50,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
110
+ pldm,pusht,50,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
111
+ pldm,pusht,50,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
112
+ pldm,pusht,75,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
113
+ pldm,pusht,75,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
114
+ pldm,pusht,75,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
115
+ pldm,pusht,75,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
116
+ pldm,pusht,75,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
117
+ pldm,pusht,100,42,0,200,,,unavailable: compatible checkpoint not found,50,,,,
118
+ pldm,pusht,100,43,0,200,,,unavailable: compatible checkpoint not found,50,,,,
119
+ pldm,pusht,100,44,0,200,,,unavailable: compatible checkpoint not found,50,,,,
120
+ pldm,pusht,100,45,0,200,,,unavailable: compatible checkpoint not found,50,,,,
121
+ pldm,pusht,100,46,0,200,,,unavailable: compatible checkpoint not found,50,,,,
122
  fast-lewm,tworoom,25,42,200,200,176,88.0,complete,50,naiverer/fast-leworldmodel,,,
123
  fast-lewm,tworoom,25,43,200,200,180,90.0,complete,50,naiverer/fast-leworldmodel,,,
124
  fast-lewm,tworoom,25,44,200,200,173,86.5,complete,50,naiverer/fast-leworldmodel,,,
results/planning_success.json CHANGED
@@ -8,7 +8,7 @@
8
  "samples_required": 200,
9
  "successes": null,
10
  "success_percent": null,
11
- "status": "queued training",
12
  "execution_budget": 50,
13
  "checkpoint_release": null,
14
  "training_epochs": null,
@@ -24,7 +24,7 @@
24
  "samples_required": 200,
25
  "successes": null,
26
  "success_percent": null,
27
- "status": "queued training",
28
  "execution_budget": 50,
29
  "checkpoint_release": null,
30
  "training_epochs": null,
@@ -40,7 +40,7 @@
40
  "samples_required": 200,
41
  "successes": null,
42
  "success_percent": null,
43
- "status": "queued training",
44
  "execution_budget": 50,
45
  "checkpoint_release": null,
46
  "training_epochs": null,
@@ -56,7 +56,7 @@
56
  "samples_required": 200,
57
  "successes": null,
58
  "success_percent": null,
59
- "status": "queued training",
60
  "execution_budget": 50,
61
  "checkpoint_release": null,
62
  "training_epochs": null,
@@ -72,7 +72,7 @@
72
  "samples_required": 200,
73
  "successes": null,
74
  "success_percent": null,
75
- "status": "queued training",
76
  "execution_budget": 50,
77
  "checkpoint_release": null,
78
  "training_epochs": null,
@@ -88,7 +88,7 @@
88
  "samples_required": 200,
89
  "successes": null,
90
  "success_percent": null,
91
- "status": "queued training",
92
  "execution_budget": 50,
93
  "checkpoint_release": null,
94
  "training_epochs": null,
@@ -104,7 +104,7 @@
104
  "samples_required": 200,
105
  "successes": null,
106
  "success_percent": null,
107
- "status": "queued training",
108
  "execution_budget": 50,
109
  "checkpoint_release": null,
110
  "training_epochs": null,
@@ -120,7 +120,7 @@
120
  "samples_required": 200,
121
  "successes": null,
122
  "success_percent": null,
123
- "status": "queued training",
124
  "execution_budget": 50,
125
  "checkpoint_release": null,
126
  "training_epochs": null,
@@ -136,7 +136,7 @@
136
  "samples_required": 200,
137
  "successes": null,
138
  "success_percent": null,
139
- "status": "queued training",
140
  "execution_budget": 50,
141
  "checkpoint_release": null,
142
  "training_epochs": null,
@@ -152,7 +152,7 @@
152
  "samples_required": 200,
153
  "successes": null,
154
  "success_percent": null,
155
- "status": "queued training",
156
  "execution_budget": 50,
157
  "checkpoint_release": null,
158
  "training_epochs": null,
@@ -168,7 +168,7 @@
168
  "samples_required": 200,
169
  "successes": null,
170
  "success_percent": null,
171
- "status": "queued training",
172
  "execution_budget": 50,
173
  "checkpoint_release": null,
174
  "training_epochs": null,
@@ -184,7 +184,7 @@
184
  "samples_required": 200,
185
  "successes": null,
186
  "success_percent": null,
187
- "status": "queued training",
188
  "execution_budget": 50,
189
  "checkpoint_release": null,
190
  "training_epochs": null,
@@ -200,7 +200,7 @@
200
  "samples_required": 200,
201
  "successes": null,
202
  "success_percent": null,
203
- "status": "queued training",
204
  "execution_budget": 50,
205
  "checkpoint_release": null,
206
  "training_epochs": null,
@@ -216,7 +216,7 @@
216
  "samples_required": 200,
217
  "successes": null,
218
  "success_percent": null,
219
- "status": "queued training",
220
  "execution_budget": 50,
221
  "checkpoint_release": null,
222
  "training_epochs": null,
@@ -232,7 +232,7 @@
232
  "samples_required": 200,
233
  "successes": null,
234
  "success_percent": null,
235
- "status": "queued training",
236
  "execution_budget": 50,
237
  "checkpoint_release": null,
238
  "training_epochs": null,
@@ -248,7 +248,7 @@
248
  "samples_required": 200,
249
  "successes": null,
250
  "success_percent": null,
251
- "status": "queued training",
252
  "execution_budget": 50,
253
  "checkpoint_release": null,
254
  "training_epochs": null,
@@ -264,7 +264,7 @@
264
  "samples_required": 200,
265
  "successes": null,
266
  "success_percent": null,
267
- "status": "queued training",
268
  "execution_budget": 50,
269
  "checkpoint_release": null,
270
  "training_epochs": null,
@@ -280,7 +280,7 @@
280
  "samples_required": 200,
281
  "successes": null,
282
  "success_percent": null,
283
- "status": "queued training",
284
  "execution_budget": 50,
285
  "checkpoint_release": null,
286
  "training_epochs": null,
@@ -296,7 +296,7 @@
296
  "samples_required": 200,
297
  "successes": null,
298
  "success_percent": null,
299
- "status": "queued training",
300
  "execution_budget": 50,
301
  "checkpoint_release": null,
302
  "training_epochs": null,
@@ -312,7 +312,7 @@
312
  "samples_required": 200,
313
  "successes": null,
314
  "success_percent": null,
315
- "status": "queued training",
316
  "execution_budget": 50,
317
  "checkpoint_release": null,
318
  "training_epochs": null,
@@ -328,7 +328,7 @@
328
  "samples_required": 200,
329
  "successes": null,
330
  "success_percent": null,
331
- "status": "queued training",
332
  "execution_budget": 50,
333
  "checkpoint_release": null,
334
  "training_epochs": null,
@@ -344,7 +344,7 @@
344
  "samples_required": 200,
345
  "successes": null,
346
  "success_percent": null,
347
- "status": "queued training",
348
  "execution_budget": 50,
349
  "checkpoint_release": null,
350
  "training_epochs": null,
@@ -360,7 +360,7 @@
360
  "samples_required": 200,
361
  "successes": null,
362
  "success_percent": null,
363
- "status": "queued training",
364
  "execution_budget": 50,
365
  "checkpoint_release": null,
366
  "training_epochs": null,
@@ -376,7 +376,7 @@
376
  "samples_required": 200,
377
  "successes": null,
378
  "success_percent": null,
379
- "status": "queued training",
380
  "execution_budget": 50,
381
  "checkpoint_release": null,
382
  "training_epochs": null,
@@ -392,7 +392,7 @@
392
  "samples_required": 200,
393
  "successes": null,
394
  "success_percent": null,
395
- "status": "queued training",
396
  "execution_budget": 50,
397
  "checkpoint_release": null,
398
  "training_epochs": null,
@@ -408,7 +408,7 @@
408
  "samples_required": 200,
409
  "successes": null,
410
  "success_percent": null,
411
- "status": "queued training",
412
  "execution_budget": 50,
413
  "checkpoint_release": null,
414
  "training_epochs": null,
@@ -424,7 +424,7 @@
424
  "samples_required": 200,
425
  "successes": null,
426
  "success_percent": null,
427
- "status": "queued training",
428
  "execution_budget": 50,
429
  "checkpoint_release": null,
430
  "training_epochs": null,
@@ -440,7 +440,7 @@
440
  "samples_required": 200,
441
  "successes": null,
442
  "success_percent": null,
443
- "status": "queued training",
444
  "execution_budget": 50,
445
  "checkpoint_release": null,
446
  "training_epochs": null,
@@ -456,7 +456,7 @@
456
  "samples_required": 200,
457
  "successes": null,
458
  "success_percent": null,
459
- "status": "queued training",
460
  "execution_budget": 50,
461
  "checkpoint_release": null,
462
  "training_epochs": null,
@@ -472,7 +472,7 @@
472
  "samples_required": 200,
473
  "successes": null,
474
  "success_percent": null,
475
- "status": "queued training",
476
  "execution_budget": 50,
477
  "checkpoint_release": null,
478
  "training_epochs": null,
@@ -488,7 +488,7 @@
488
  "samples_required": 200,
489
  "successes": null,
490
  "success_percent": null,
491
- "status": "queued training",
492
  "execution_budget": 50,
493
  "checkpoint_release": null,
494
  "training_epochs": null,
@@ -504,7 +504,7 @@
504
  "samples_required": 200,
505
  "successes": null,
506
  "success_percent": null,
507
- "status": "queued training",
508
  "execution_budget": 50,
509
  "checkpoint_release": null,
510
  "training_epochs": null,
@@ -520,7 +520,7 @@
520
  "samples_required": 200,
521
  "successes": null,
522
  "success_percent": null,
523
- "status": "queued training",
524
  "execution_budget": 50,
525
  "checkpoint_release": null,
526
  "training_epochs": null,
@@ -536,7 +536,7 @@
536
  "samples_required": 200,
537
  "successes": null,
538
  "success_percent": null,
539
- "status": "queued training",
540
  "execution_budget": 50,
541
  "checkpoint_release": null,
542
  "training_epochs": null,
@@ -552,7 +552,7 @@
552
  "samples_required": 200,
553
  "successes": null,
554
  "success_percent": null,
555
- "status": "queued training",
556
  "execution_budget": 50,
557
  "checkpoint_release": null,
558
  "training_epochs": null,
@@ -568,7 +568,7 @@
568
  "samples_required": 200,
569
  "successes": null,
570
  "success_percent": null,
571
- "status": "queued training",
572
  "execution_budget": 50,
573
  "checkpoint_release": null,
574
  "training_epochs": null,
@@ -584,7 +584,7 @@
584
  "samples_required": 200,
585
  "successes": null,
586
  "success_percent": null,
587
- "status": "queued training",
588
  "execution_budget": 50,
589
  "checkpoint_release": null,
590
  "training_epochs": null,
@@ -600,7 +600,7 @@
600
  "samples_required": 200,
601
  "successes": null,
602
  "success_percent": null,
603
- "status": "queued training",
604
  "execution_budget": 50,
605
  "checkpoint_release": null,
606
  "training_epochs": null,
@@ -616,7 +616,7 @@
616
  "samples_required": 200,
617
  "successes": null,
618
  "success_percent": null,
619
- "status": "queued training",
620
  "execution_budget": 50,
621
  "checkpoint_release": null,
622
  "training_epochs": null,
@@ -632,7 +632,7 @@
632
  "samples_required": 200,
633
  "successes": null,
634
  "success_percent": null,
635
- "status": "queued training",
636
  "execution_budget": 50,
637
  "checkpoint_release": null,
638
  "training_epochs": null,
@@ -1284,320 +1284,320 @@
1284
  "dataset": "cube",
1285
  "offset": 25,
1286
  "seed": 42,
1287
- "samples_completed": 0,
1288
  "samples_required": 200,
1289
- "successes": null,
1290
- "success_percent": null,
1291
- "status": "training",
1292
  "execution_budget": 50,
1293
- "checkpoint_release": null,
1294
- "training_epochs": null,
1295
- "training_seed": null,
1296
- "model_variant": null
1297
  },
1298
  {
1299
  "model": "pldm",
1300
  "dataset": "cube",
1301
  "offset": 25,
1302
  "seed": 43,
1303
- "samples_completed": 0,
1304
  "samples_required": 200,
1305
- "successes": null,
1306
- "success_percent": null,
1307
- "status": "training",
1308
  "execution_budget": 50,
1309
- "checkpoint_release": null,
1310
- "training_epochs": null,
1311
- "training_seed": null,
1312
- "model_variant": null
1313
  },
1314
  {
1315
  "model": "pldm",
1316
  "dataset": "cube",
1317
  "offset": 25,
1318
  "seed": 44,
1319
- "samples_completed": 0,
1320
  "samples_required": 200,
1321
- "successes": null,
1322
- "success_percent": null,
1323
- "status": "training",
1324
  "execution_budget": 50,
1325
- "checkpoint_release": null,
1326
- "training_epochs": null,
1327
- "training_seed": null,
1328
- "model_variant": null
1329
  },
1330
  {
1331
  "model": "pldm",
1332
  "dataset": "cube",
1333
  "offset": 25,
1334
  "seed": 45,
1335
- "samples_completed": 0,
1336
  "samples_required": 200,
1337
- "successes": null,
1338
- "success_percent": null,
1339
- "status": "training",
1340
  "execution_budget": 50,
1341
- "checkpoint_release": null,
1342
- "training_epochs": null,
1343
- "training_seed": null,
1344
- "model_variant": null
1345
  },
1346
  {
1347
  "model": "pldm",
1348
  "dataset": "cube",
1349
  "offset": 25,
1350
  "seed": 46,
1351
- "samples_completed": 0,
1352
  "samples_required": 200,
1353
- "successes": null,
1354
- "success_percent": null,
1355
- "status": "training",
1356
  "execution_budget": 50,
1357
- "checkpoint_release": null,
1358
- "training_epochs": null,
1359
- "training_seed": null,
1360
- "model_variant": null
1361
  },
1362
  {
1363
  "model": "pldm",
1364
  "dataset": "cube",
1365
  "offset": 50,
1366
  "seed": 42,
1367
- "samples_completed": 0,
1368
  "samples_required": 200,
1369
- "successes": null,
1370
- "success_percent": null,
1371
- "status": "training",
1372
  "execution_budget": 50,
1373
- "checkpoint_release": null,
1374
- "training_epochs": null,
1375
- "training_seed": null,
1376
- "model_variant": null
1377
  },
1378
  {
1379
  "model": "pldm",
1380
  "dataset": "cube",
1381
  "offset": 50,
1382
  "seed": 43,
1383
- "samples_completed": 0,
1384
  "samples_required": 200,
1385
- "successes": null,
1386
- "success_percent": null,
1387
- "status": "training",
1388
  "execution_budget": 50,
1389
- "checkpoint_release": null,
1390
- "training_epochs": null,
1391
- "training_seed": null,
1392
- "model_variant": null
1393
  },
1394
  {
1395
  "model": "pldm",
1396
  "dataset": "cube",
1397
  "offset": 50,
1398
  "seed": 44,
1399
- "samples_completed": 0,
1400
  "samples_required": 200,
1401
- "successes": null,
1402
- "success_percent": null,
1403
- "status": "training",
1404
  "execution_budget": 50,
1405
- "checkpoint_release": null,
1406
- "training_epochs": null,
1407
- "training_seed": null,
1408
- "model_variant": null
1409
  },
1410
  {
1411
  "model": "pldm",
1412
  "dataset": "cube",
1413
  "offset": 50,
1414
  "seed": 45,
1415
- "samples_completed": 0,
1416
  "samples_required": 200,
1417
- "successes": null,
1418
- "success_percent": null,
1419
- "status": "training",
1420
  "execution_budget": 50,
1421
- "checkpoint_release": null,
1422
- "training_epochs": null,
1423
- "training_seed": null,
1424
- "model_variant": null
1425
  },
1426
  {
1427
  "model": "pldm",
1428
  "dataset": "cube",
1429
  "offset": 50,
1430
  "seed": 46,
1431
- "samples_completed": 0,
1432
  "samples_required": 200,
1433
- "successes": null,
1434
- "success_percent": null,
1435
- "status": "training",
1436
  "execution_budget": 50,
1437
- "checkpoint_release": null,
1438
- "training_epochs": null,
1439
- "training_seed": null,
1440
- "model_variant": null
1441
  },
1442
  {
1443
  "model": "pldm",
1444
  "dataset": "cube",
1445
  "offset": 75,
1446
  "seed": 42,
1447
- "samples_completed": 0,
1448
  "samples_required": 200,
1449
- "successes": null,
1450
- "success_percent": null,
1451
- "status": "training",
1452
  "execution_budget": 50,
1453
- "checkpoint_release": null,
1454
- "training_epochs": null,
1455
- "training_seed": null,
1456
- "model_variant": null
1457
  },
1458
  {
1459
  "model": "pldm",
1460
  "dataset": "cube",
1461
  "offset": 75,
1462
  "seed": 43,
1463
- "samples_completed": 0,
1464
  "samples_required": 200,
1465
- "successes": null,
1466
- "success_percent": null,
1467
- "status": "training",
1468
  "execution_budget": 50,
1469
- "checkpoint_release": null,
1470
- "training_epochs": null,
1471
- "training_seed": null,
1472
- "model_variant": null
1473
  },
1474
  {
1475
  "model": "pldm",
1476
  "dataset": "cube",
1477
  "offset": 75,
1478
  "seed": 44,
1479
- "samples_completed": 0,
1480
  "samples_required": 200,
1481
- "successes": null,
1482
- "success_percent": null,
1483
- "status": "training",
1484
  "execution_budget": 50,
1485
- "checkpoint_release": null,
1486
- "training_epochs": null,
1487
- "training_seed": null,
1488
- "model_variant": null
1489
  },
1490
  {
1491
  "model": "pldm",
1492
  "dataset": "cube",
1493
  "offset": 75,
1494
  "seed": 45,
1495
- "samples_completed": 0,
1496
  "samples_required": 200,
1497
- "successes": null,
1498
- "success_percent": null,
1499
- "status": "training",
1500
  "execution_budget": 50,
1501
- "checkpoint_release": null,
1502
- "training_epochs": null,
1503
- "training_seed": null,
1504
- "model_variant": null
1505
  },
1506
  {
1507
  "model": "pldm",
1508
  "dataset": "cube",
1509
  "offset": 75,
1510
  "seed": 46,
1511
- "samples_completed": 0,
1512
  "samples_required": 200,
1513
- "successes": null,
1514
- "success_percent": null,
1515
- "status": "training",
1516
  "execution_budget": 50,
1517
- "checkpoint_release": null,
1518
- "training_epochs": null,
1519
- "training_seed": null,
1520
- "model_variant": null
1521
  },
1522
  {
1523
  "model": "pldm",
1524
  "dataset": "cube",
1525
  "offset": 100,
1526
  "seed": 42,
1527
- "samples_completed": 0,
1528
  "samples_required": 200,
1529
- "successes": null,
1530
- "success_percent": null,
1531
- "status": "training",
1532
  "execution_budget": 50,
1533
- "checkpoint_release": null,
1534
- "training_epochs": null,
1535
- "training_seed": null,
1536
- "model_variant": null
1537
  },
1538
  {
1539
  "model": "pldm",
1540
  "dataset": "cube",
1541
  "offset": 100,
1542
  "seed": 43,
1543
- "samples_completed": 0,
1544
  "samples_required": 200,
1545
- "successes": null,
1546
- "success_percent": null,
1547
- "status": "training",
1548
  "execution_budget": 50,
1549
- "checkpoint_release": null,
1550
- "training_epochs": null,
1551
- "training_seed": null,
1552
- "model_variant": null
1553
  },
1554
  {
1555
  "model": "pldm",
1556
  "dataset": "cube",
1557
  "offset": 100,
1558
  "seed": 44,
1559
- "samples_completed": 0,
1560
  "samples_required": 200,
1561
- "successes": null,
1562
- "success_percent": null,
1563
- "status": "training",
1564
  "execution_budget": 50,
1565
- "checkpoint_release": null,
1566
- "training_epochs": null,
1567
- "training_seed": null,
1568
- "model_variant": null
1569
  },
1570
  {
1571
  "model": "pldm",
1572
  "dataset": "cube",
1573
  "offset": 100,
1574
  "seed": 45,
1575
- "samples_completed": 0,
1576
  "samples_required": 200,
1577
- "successes": null,
1578
- "success_percent": null,
1579
- "status": "training",
1580
  "execution_budget": 50,
1581
- "checkpoint_release": null,
1582
- "training_epochs": null,
1583
- "training_seed": null,
1584
- "model_variant": null
1585
  },
1586
  {
1587
  "model": "pldm",
1588
  "dataset": "cube",
1589
  "offset": 100,
1590
  "seed": 46,
1591
- "samples_completed": 0,
1592
  "samples_required": 200,
1593
- "successes": null,
1594
- "success_percent": null,
1595
- "status": "training",
1596
  "execution_budget": 50,
1597
- "checkpoint_release": null,
1598
- "training_epochs": null,
1599
- "training_seed": null,
1600
- "model_variant": null
1601
  },
1602
  {
1603
  "model": "pldm",
@@ -1608,7 +1608,7 @@
1608
  "samples_required": 200,
1609
  "successes": null,
1610
  "success_percent": null,
1611
- "status": "queued training",
1612
  "execution_budget": 50,
1613
  "checkpoint_release": null,
1614
  "training_epochs": null,
@@ -1624,7 +1624,7 @@
1624
  "samples_required": 200,
1625
  "successes": null,
1626
  "success_percent": null,
1627
- "status": "queued training",
1628
  "execution_budget": 50,
1629
  "checkpoint_release": null,
1630
  "training_epochs": null,
@@ -1640,7 +1640,7 @@
1640
  "samples_required": 200,
1641
  "successes": null,
1642
  "success_percent": null,
1643
- "status": "queued training",
1644
  "execution_budget": 50,
1645
  "checkpoint_release": null,
1646
  "training_epochs": null,
@@ -1656,7 +1656,7 @@
1656
  "samples_required": 200,
1657
  "successes": null,
1658
  "success_percent": null,
1659
- "status": "queued training",
1660
  "execution_budget": 50,
1661
  "checkpoint_release": null,
1662
  "training_epochs": null,
@@ -1672,7 +1672,7 @@
1672
  "samples_required": 200,
1673
  "successes": null,
1674
  "success_percent": null,
1675
- "status": "queued training",
1676
  "execution_budget": 50,
1677
  "checkpoint_release": null,
1678
  "training_epochs": null,
@@ -1688,7 +1688,7 @@
1688
  "samples_required": 200,
1689
  "successes": null,
1690
  "success_percent": null,
1691
- "status": "queued training",
1692
  "execution_budget": 50,
1693
  "checkpoint_release": null,
1694
  "training_epochs": null,
@@ -1704,7 +1704,7 @@
1704
  "samples_required": 200,
1705
  "successes": null,
1706
  "success_percent": null,
1707
- "status": "queued training",
1708
  "execution_budget": 50,
1709
  "checkpoint_release": null,
1710
  "training_epochs": null,
@@ -1720,7 +1720,7 @@
1720
  "samples_required": 200,
1721
  "successes": null,
1722
  "success_percent": null,
1723
- "status": "queued training",
1724
  "execution_budget": 50,
1725
  "checkpoint_release": null,
1726
  "training_epochs": null,
@@ -1736,7 +1736,7 @@
1736
  "samples_required": 200,
1737
  "successes": null,
1738
  "success_percent": null,
1739
- "status": "queued training",
1740
  "execution_budget": 50,
1741
  "checkpoint_release": null,
1742
  "training_epochs": null,
@@ -1752,7 +1752,7 @@
1752
  "samples_required": 200,
1753
  "successes": null,
1754
  "success_percent": null,
1755
- "status": "queued training",
1756
  "execution_budget": 50,
1757
  "checkpoint_release": null,
1758
  "training_epochs": null,
@@ -1768,7 +1768,7 @@
1768
  "samples_required": 200,
1769
  "successes": null,
1770
  "success_percent": null,
1771
- "status": "queued training",
1772
  "execution_budget": 50,
1773
  "checkpoint_release": null,
1774
  "training_epochs": null,
@@ -1784,7 +1784,7 @@
1784
  "samples_required": 200,
1785
  "successes": null,
1786
  "success_percent": null,
1787
- "status": "queued training",
1788
  "execution_budget": 50,
1789
  "checkpoint_release": null,
1790
  "training_epochs": null,
@@ -1800,7 +1800,7 @@
1800
  "samples_required": 200,
1801
  "successes": null,
1802
  "success_percent": null,
1803
- "status": "queued training",
1804
  "execution_budget": 50,
1805
  "checkpoint_release": null,
1806
  "training_epochs": null,
@@ -1816,7 +1816,7 @@
1816
  "samples_required": 200,
1817
  "successes": null,
1818
  "success_percent": null,
1819
- "status": "queued training",
1820
  "execution_budget": 50,
1821
  "checkpoint_release": null,
1822
  "training_epochs": null,
@@ -1832,7 +1832,7 @@
1832
  "samples_required": 200,
1833
  "successes": null,
1834
  "success_percent": null,
1835
- "status": "queued training",
1836
  "execution_budget": 50,
1837
  "checkpoint_release": null,
1838
  "training_epochs": null,
@@ -1848,7 +1848,7 @@
1848
  "samples_required": 200,
1849
  "successes": null,
1850
  "success_percent": null,
1851
- "status": "queued training",
1852
  "execution_budget": 50,
1853
  "checkpoint_release": null,
1854
  "training_epochs": null,
@@ -1864,7 +1864,7 @@
1864
  "samples_required": 200,
1865
  "successes": null,
1866
  "success_percent": null,
1867
- "status": "queued training",
1868
  "execution_budget": 50,
1869
  "checkpoint_release": null,
1870
  "training_epochs": null,
@@ -1880,7 +1880,7 @@
1880
  "samples_required": 200,
1881
  "successes": null,
1882
  "success_percent": null,
1883
- "status": "queued training",
1884
  "execution_budget": 50,
1885
  "checkpoint_release": null,
1886
  "training_epochs": null,
@@ -1896,7 +1896,7 @@
1896
  "samples_required": 200,
1897
  "successes": null,
1898
  "success_percent": null,
1899
- "status": "queued training",
1900
  "execution_budget": 50,
1901
  "checkpoint_release": null,
1902
  "training_epochs": null,
@@ -1912,7 +1912,7 @@
1912
  "samples_required": 200,
1913
  "successes": null,
1914
  "success_percent": null,
1915
- "status": "queued training",
1916
  "execution_budget": 50,
1917
  "checkpoint_release": null,
1918
  "training_epochs": null,
 
8
  "samples_required": 200,
9
  "successes": null,
10
  "success_percent": null,
11
+ "status": "unavailable: compatible checkpoint not found",
12
  "execution_budget": 50,
13
  "checkpoint_release": null,
14
  "training_epochs": null,
 
24
  "samples_required": 200,
25
  "successes": null,
26
  "success_percent": null,
27
+ "status": "unavailable: compatible checkpoint not found",
28
  "execution_budget": 50,
29
  "checkpoint_release": null,
30
  "training_epochs": null,
 
40
  "samples_required": 200,
41
  "successes": null,
42
  "success_percent": null,
43
+ "status": "unavailable: compatible checkpoint not found",
44
  "execution_budget": 50,
45
  "checkpoint_release": null,
46
  "training_epochs": null,
 
56
  "samples_required": 200,
57
  "successes": null,
58
  "success_percent": null,
59
+ "status": "unavailable: compatible checkpoint not found",
60
  "execution_budget": 50,
61
  "checkpoint_release": null,
62
  "training_epochs": null,
 
72
  "samples_required": 200,
73
  "successes": null,
74
  "success_percent": null,
75
+ "status": "unavailable: compatible checkpoint not found",
76
  "execution_budget": 50,
77
  "checkpoint_release": null,
78
  "training_epochs": null,
 
88
  "samples_required": 200,
89
  "successes": null,
90
  "success_percent": null,
91
+ "status": "unavailable: compatible checkpoint not found",
92
  "execution_budget": 50,
93
  "checkpoint_release": null,
94
  "training_epochs": null,
 
104
  "samples_required": 200,
105
  "successes": null,
106
  "success_percent": null,
107
+ "status": "unavailable: compatible checkpoint not found",
108
  "execution_budget": 50,
109
  "checkpoint_release": null,
110
  "training_epochs": null,
 
120
  "samples_required": 200,
121
  "successes": null,
122
  "success_percent": null,
123
+ "status": "unavailable: compatible checkpoint not found",
124
  "execution_budget": 50,
125
  "checkpoint_release": null,
126
  "training_epochs": null,
 
136
  "samples_required": 200,
137
  "successes": null,
138
  "success_percent": null,
139
+ "status": "unavailable: compatible checkpoint not found",
140
  "execution_budget": 50,
141
  "checkpoint_release": null,
142
  "training_epochs": null,
 
152
  "samples_required": 200,
153
  "successes": null,
154
  "success_percent": null,
155
+ "status": "unavailable: compatible checkpoint not found",
156
  "execution_budget": 50,
157
  "checkpoint_release": null,
158
  "training_epochs": null,
 
168
  "samples_required": 200,
169
  "successes": null,
170
  "success_percent": null,
171
+ "status": "unavailable: compatible checkpoint not found",
172
  "execution_budget": 50,
173
  "checkpoint_release": null,
174
  "training_epochs": null,
 
184
  "samples_required": 200,
185
  "successes": null,
186
  "success_percent": null,
187
+ "status": "unavailable: compatible checkpoint not found",
188
  "execution_budget": 50,
189
  "checkpoint_release": null,
190
  "training_epochs": null,
 
200
  "samples_required": 200,
201
  "successes": null,
202
  "success_percent": null,
203
+ "status": "unavailable: compatible checkpoint not found",
204
  "execution_budget": 50,
205
  "checkpoint_release": null,
206
  "training_epochs": null,
 
216
  "samples_required": 200,
217
  "successes": null,
218
  "success_percent": null,
219
+ "status": "unavailable: compatible checkpoint not found",
220
  "execution_budget": 50,
221
  "checkpoint_release": null,
222
  "training_epochs": null,
 
232
  "samples_required": 200,
233
  "successes": null,
234
  "success_percent": null,
235
+ "status": "unavailable: compatible checkpoint not found",
236
  "execution_budget": 50,
237
  "checkpoint_release": null,
238
  "training_epochs": null,
 
248
  "samples_required": 200,
249
  "successes": null,
250
  "success_percent": null,
251
+ "status": "unavailable: compatible checkpoint not found",
252
  "execution_budget": 50,
253
  "checkpoint_release": null,
254
  "training_epochs": null,
 
264
  "samples_required": 200,
265
  "successes": null,
266
  "success_percent": null,
267
+ "status": "unavailable: compatible checkpoint not found",
268
  "execution_budget": 50,
269
  "checkpoint_release": null,
270
  "training_epochs": null,
 
280
  "samples_required": 200,
281
  "successes": null,
282
  "success_percent": null,
283
+ "status": "unavailable: compatible checkpoint not found",
284
  "execution_budget": 50,
285
  "checkpoint_release": null,
286
  "training_epochs": null,
 
296
  "samples_required": 200,
297
  "successes": null,
298
  "success_percent": null,
299
+ "status": "unavailable: compatible checkpoint not found",
300
  "execution_budget": 50,
301
  "checkpoint_release": null,
302
  "training_epochs": null,
 
312
  "samples_required": 200,
313
  "successes": null,
314
  "success_percent": null,
315
+ "status": "unavailable: compatible checkpoint not found",
316
  "execution_budget": 50,
317
  "checkpoint_release": null,
318
  "training_epochs": null,
 
328
  "samples_required": 200,
329
  "successes": null,
330
  "success_percent": null,
331
+ "status": "unavailable: compatible checkpoint not found",
332
  "execution_budget": 50,
333
  "checkpoint_release": null,
334
  "training_epochs": null,
 
344
  "samples_required": 200,
345
  "successes": null,
346
  "success_percent": null,
347
+ "status": "unavailable: compatible checkpoint not found",
348
  "execution_budget": 50,
349
  "checkpoint_release": null,
350
  "training_epochs": null,
 
360
  "samples_required": 200,
361
  "successes": null,
362
  "success_percent": null,
363
+ "status": "unavailable: compatible checkpoint not found",
364
  "execution_budget": 50,
365
  "checkpoint_release": null,
366
  "training_epochs": null,
 
376
  "samples_required": 200,
377
  "successes": null,
378
  "success_percent": null,
379
+ "status": "unavailable: compatible checkpoint not found",
380
  "execution_budget": 50,
381
  "checkpoint_release": null,
382
  "training_epochs": null,
 
392
  "samples_required": 200,
393
  "successes": null,
394
  "success_percent": null,
395
+ "status": "unavailable: compatible checkpoint not found",
396
  "execution_budget": 50,
397
  "checkpoint_release": null,
398
  "training_epochs": null,
 
408
  "samples_required": 200,
409
  "successes": null,
410
  "success_percent": null,
411
+ "status": "unavailable: compatible checkpoint not found",
412
  "execution_budget": 50,
413
  "checkpoint_release": null,
414
  "training_epochs": null,
 
424
  "samples_required": 200,
425
  "successes": null,
426
  "success_percent": null,
427
+ "status": "unavailable: compatible checkpoint not found",
428
  "execution_budget": 50,
429
  "checkpoint_release": null,
430
  "training_epochs": null,
 
440
  "samples_required": 200,
441
  "successes": null,
442
  "success_percent": null,
443
+ "status": "unavailable: compatible checkpoint not found",
444
  "execution_budget": 50,
445
  "checkpoint_release": null,
446
  "training_epochs": null,
 
456
  "samples_required": 200,
457
  "successes": null,
458
  "success_percent": null,
459
+ "status": "unavailable: compatible checkpoint not found",
460
  "execution_budget": 50,
461
  "checkpoint_release": null,
462
  "training_epochs": null,
 
472
  "samples_required": 200,
473
  "successes": null,
474
  "success_percent": null,
475
+ "status": "unavailable: compatible checkpoint not found",
476
  "execution_budget": 50,
477
  "checkpoint_release": null,
478
  "training_epochs": null,
 
488
  "samples_required": 200,
489
  "successes": null,
490
  "success_percent": null,
491
+ "status": "unavailable: compatible checkpoint not found",
492
  "execution_budget": 50,
493
  "checkpoint_release": null,
494
  "training_epochs": null,
 
504
  "samples_required": 200,
505
  "successes": null,
506
  "success_percent": null,
507
+ "status": "unavailable: compatible checkpoint not found",
508
  "execution_budget": 50,
509
  "checkpoint_release": null,
510
  "training_epochs": null,
 
520
  "samples_required": 200,
521
  "successes": null,
522
  "success_percent": null,
523
+ "status": "unavailable: compatible checkpoint not found",
524
  "execution_budget": 50,
525
  "checkpoint_release": null,
526
  "training_epochs": null,
 
536
  "samples_required": 200,
537
  "successes": null,
538
  "success_percent": null,
539
+ "status": "unavailable: compatible checkpoint not found",
540
  "execution_budget": 50,
541
  "checkpoint_release": null,
542
  "training_epochs": null,
 
552
  "samples_required": 200,
553
  "successes": null,
554
  "success_percent": null,
555
+ "status": "unavailable: compatible checkpoint not found",
556
  "execution_budget": 50,
557
  "checkpoint_release": null,
558
  "training_epochs": null,
 
568
  "samples_required": 200,
569
  "successes": null,
570
  "success_percent": null,
571
+ "status": "unavailable: compatible checkpoint not found",
572
  "execution_budget": 50,
573
  "checkpoint_release": null,
574
  "training_epochs": null,
 
584
  "samples_required": 200,
585
  "successes": null,
586
  "success_percent": null,
587
+ "status": "unavailable: compatible checkpoint not found",
588
  "execution_budget": 50,
589
  "checkpoint_release": null,
590
  "training_epochs": null,
 
600
  "samples_required": 200,
601
  "successes": null,
602
  "success_percent": null,
603
+ "status": "unavailable: compatible checkpoint not found",
604
  "execution_budget": 50,
605
  "checkpoint_release": null,
606
  "training_epochs": null,
 
616
  "samples_required": 200,
617
  "successes": null,
618
  "success_percent": null,
619
+ "status": "unavailable: compatible checkpoint not found",
620
  "execution_budget": 50,
621
  "checkpoint_release": null,
622
  "training_epochs": null,
 
632
  "samples_required": 200,
633
  "successes": null,
634
  "success_percent": null,
635
+ "status": "unavailable: compatible checkpoint not found",
636
  "execution_budget": 50,
637
  "checkpoint_release": null,
638
  "training_epochs": null,
 
1284
  "dataset": "cube",
1285
  "offset": 25,
1286
  "seed": 42,
1287
+ "samples_completed": 200,
1288
  "samples_required": 200,
1289
+ "successes": 110,
1290
+ "success_percent": 55.0,
1291
+ "status": "complete",
1292
  "execution_budget": 50,
1293
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1294
+ "training_epochs": 10,
1295
+ "training_seed": 42,
1296
+ "model_variant": "PLDM stable-worldmodel recipe"
1297
  },
1298
  {
1299
  "model": "pldm",
1300
  "dataset": "cube",
1301
  "offset": 25,
1302
  "seed": 43,
1303
+ "samples_completed": 200,
1304
  "samples_required": 200,
1305
+ "successes": 104,
1306
+ "success_percent": 52.0,
1307
+ "status": "complete",
1308
  "execution_budget": 50,
1309
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1310
+ "training_epochs": 10,
1311
+ "training_seed": 42,
1312
+ "model_variant": "PLDM stable-worldmodel recipe"
1313
  },
1314
  {
1315
  "model": "pldm",
1316
  "dataset": "cube",
1317
  "offset": 25,
1318
  "seed": 44,
1319
+ "samples_completed": 200,
1320
  "samples_required": 200,
1321
+ "successes": 112,
1322
+ "success_percent": 56.0,
1323
+ "status": "complete",
1324
  "execution_budget": 50,
1325
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1326
+ "training_epochs": 10,
1327
+ "training_seed": 42,
1328
+ "model_variant": "PLDM stable-worldmodel recipe"
1329
  },
1330
  {
1331
  "model": "pldm",
1332
  "dataset": "cube",
1333
  "offset": 25,
1334
  "seed": 45,
1335
+ "samples_completed": 200,
1336
  "samples_required": 200,
1337
+ "successes": 113,
1338
+ "success_percent": 56.5,
1339
+ "status": "complete",
1340
  "execution_budget": 50,
1341
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1342
+ "training_epochs": 10,
1343
+ "training_seed": 42,
1344
+ "model_variant": "PLDM stable-worldmodel recipe"
1345
  },
1346
  {
1347
  "model": "pldm",
1348
  "dataset": "cube",
1349
  "offset": 25,
1350
  "seed": 46,
1351
+ "samples_completed": 200,
1352
  "samples_required": 200,
1353
+ "successes": 132,
1354
+ "success_percent": 66.0,
1355
+ "status": "complete",
1356
  "execution_budget": 50,
1357
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1358
+ "training_epochs": 10,
1359
+ "training_seed": 42,
1360
+ "model_variant": "PLDM stable-worldmodel recipe"
1361
  },
1362
  {
1363
  "model": "pldm",
1364
  "dataset": "cube",
1365
  "offset": 50,
1366
  "seed": 42,
1367
+ "samples_completed": 200,
1368
  "samples_required": 200,
1369
+ "successes": 99,
1370
+ "success_percent": 49.5,
1371
+ "status": "complete",
1372
  "execution_budget": 50,
1373
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1374
+ "training_epochs": 10,
1375
+ "training_seed": 42,
1376
+ "model_variant": "PLDM stable-worldmodel recipe"
1377
  },
1378
  {
1379
  "model": "pldm",
1380
  "dataset": "cube",
1381
  "offset": 50,
1382
  "seed": 43,
1383
+ "samples_completed": 200,
1384
  "samples_required": 200,
1385
+ "successes": 78,
1386
+ "success_percent": 39.0,
1387
+ "status": "complete",
1388
  "execution_budget": 50,
1389
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1390
+ "training_epochs": 10,
1391
+ "training_seed": 42,
1392
+ "model_variant": "PLDM stable-worldmodel recipe"
1393
  },
1394
  {
1395
  "model": "pldm",
1396
  "dataset": "cube",
1397
  "offset": 50,
1398
  "seed": 44,
1399
+ "samples_completed": 200,
1400
  "samples_required": 200,
1401
+ "successes": 83,
1402
+ "success_percent": 41.5,
1403
+ "status": "complete",
1404
  "execution_budget": 50,
1405
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1406
+ "training_epochs": 10,
1407
+ "training_seed": 42,
1408
+ "model_variant": "PLDM stable-worldmodel recipe"
1409
  },
1410
  {
1411
  "model": "pldm",
1412
  "dataset": "cube",
1413
  "offset": 50,
1414
  "seed": 45,
1415
+ "samples_completed": 200,
1416
  "samples_required": 200,
1417
+ "successes": 89,
1418
+ "success_percent": 44.5,
1419
+ "status": "complete",
1420
  "execution_budget": 50,
1421
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1422
+ "training_epochs": 10,
1423
+ "training_seed": 42,
1424
+ "model_variant": "PLDM stable-worldmodel recipe"
1425
  },
1426
  {
1427
  "model": "pldm",
1428
  "dataset": "cube",
1429
  "offset": 50,
1430
  "seed": 46,
1431
+ "samples_completed": 200,
1432
  "samples_required": 200,
1433
+ "successes": 101,
1434
+ "success_percent": 50.5,
1435
+ "status": "complete",
1436
  "execution_budget": 50,
1437
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1438
+ "training_epochs": 10,
1439
+ "training_seed": 42,
1440
+ "model_variant": "PLDM stable-worldmodel recipe"
1441
  },
1442
  {
1443
  "model": "pldm",
1444
  "dataset": "cube",
1445
  "offset": 75,
1446
  "seed": 42,
1447
+ "samples_completed": 200,
1448
  "samples_required": 200,
1449
+ "successes": 109,
1450
+ "success_percent": 54.5,
1451
+ "status": "complete",
1452
  "execution_budget": 50,
1453
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1454
+ "training_epochs": 10,
1455
+ "training_seed": 42,
1456
+ "model_variant": "PLDM stable-worldmodel recipe"
1457
  },
1458
  {
1459
  "model": "pldm",
1460
  "dataset": "cube",
1461
  "offset": 75,
1462
  "seed": 43,
1463
+ "samples_completed": 200,
1464
  "samples_required": 200,
1465
+ "successes": 117,
1466
+ "success_percent": 58.5,
1467
+ "status": "complete",
1468
  "execution_budget": 50,
1469
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1470
+ "training_epochs": 10,
1471
+ "training_seed": 42,
1472
+ "model_variant": "PLDM stable-worldmodel recipe"
1473
  },
1474
  {
1475
  "model": "pldm",
1476
  "dataset": "cube",
1477
  "offset": 75,
1478
  "seed": 44,
1479
+ "samples_completed": 200,
1480
  "samples_required": 200,
1481
+ "successes": 114,
1482
+ "success_percent": 57.0,
1483
+ "status": "complete",
1484
  "execution_budget": 50,
1485
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1486
+ "training_epochs": 10,
1487
+ "training_seed": 42,
1488
+ "model_variant": "PLDM stable-worldmodel recipe"
1489
  },
1490
  {
1491
  "model": "pldm",
1492
  "dataset": "cube",
1493
  "offset": 75,
1494
  "seed": 45,
1495
+ "samples_completed": 200,
1496
  "samples_required": 200,
1497
+ "successes": 109,
1498
+ "success_percent": 54.5,
1499
+ "status": "complete",
1500
  "execution_budget": 50,
1501
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1502
+ "training_epochs": 10,
1503
+ "training_seed": 42,
1504
+ "model_variant": "PLDM stable-worldmodel recipe"
1505
  },
1506
  {
1507
  "model": "pldm",
1508
  "dataset": "cube",
1509
  "offset": 75,
1510
  "seed": 46,
1511
+ "samples_completed": 200,
1512
  "samples_required": 200,
1513
+ "successes": 103,
1514
+ "success_percent": 51.5,
1515
+ "status": "complete",
1516
  "execution_budget": 50,
1517
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1518
+ "training_epochs": 10,
1519
+ "training_seed": 42,
1520
+ "model_variant": "PLDM stable-worldmodel recipe"
1521
  },
1522
  {
1523
  "model": "pldm",
1524
  "dataset": "cube",
1525
  "offset": 100,
1526
  "seed": 42,
1527
+ "samples_completed": 200,
1528
  "samples_required": 200,
1529
+ "successes": 115,
1530
+ "success_percent": 57.5,
1531
+ "status": "complete",
1532
  "execution_budget": 50,
1533
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1534
+ "training_epochs": 10,
1535
+ "training_seed": 42,
1536
+ "model_variant": "PLDM stable-worldmodel recipe"
1537
  },
1538
  {
1539
  "model": "pldm",
1540
  "dataset": "cube",
1541
  "offset": 100,
1542
  "seed": 43,
1543
+ "samples_completed": 200,
1544
  "samples_required": 200,
1545
+ "successes": 116,
1546
+ "success_percent": 58.0,
1547
+ "status": "complete",
1548
  "execution_budget": 50,
1549
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1550
+ "training_epochs": 10,
1551
+ "training_seed": 42,
1552
+ "model_variant": "PLDM stable-worldmodel recipe"
1553
  },
1554
  {
1555
  "model": "pldm",
1556
  "dataset": "cube",
1557
  "offset": 100,
1558
  "seed": 44,
1559
+ "samples_completed": 200,
1560
  "samples_required": 200,
1561
+ "successes": 103,
1562
+ "success_percent": 51.5,
1563
+ "status": "complete",
1564
  "execution_budget": 50,
1565
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1566
+ "training_epochs": 10,
1567
+ "training_seed": 42,
1568
+ "model_variant": "PLDM stable-worldmodel recipe"
1569
  },
1570
  {
1571
  "model": "pldm",
1572
  "dataset": "cube",
1573
  "offset": 100,
1574
  "seed": 45,
1575
+ "samples_completed": 200,
1576
  "samples_required": 200,
1577
+ "successes": 108,
1578
+ "success_percent": 54.0,
1579
+ "status": "complete",
1580
  "execution_budget": 50,
1581
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1582
+ "training_epochs": 10,
1583
+ "training_seed": 42,
1584
+ "model_variant": "PLDM stable-worldmodel recipe"
1585
  },
1586
  {
1587
  "model": "pldm",
1588
  "dataset": "cube",
1589
  "offset": 100,
1590
  "seed": 46,
1591
+ "samples_completed": 200,
1592
  "samples_required": 200,
1593
+ "successes": 106,
1594
+ "success_percent": 53.0,
1595
+ "status": "complete",
1596
  "execution_budget": 50,
1597
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
1598
+ "training_epochs": 10,
1599
+ "training_seed": 42,
1600
+ "model_variant": "PLDM stable-worldmodel recipe"
1601
  },
1602
  {
1603
  "model": "pldm",
 
1608
  "samples_required": 200,
1609
  "successes": null,
1610
  "success_percent": null,
1611
+ "status": "unavailable: compatible checkpoint not found",
1612
  "execution_budget": 50,
1613
  "checkpoint_release": null,
1614
  "training_epochs": null,
 
1624
  "samples_required": 200,
1625
  "successes": null,
1626
  "success_percent": null,
1627
+ "status": "unavailable: compatible checkpoint not found",
1628
  "execution_budget": 50,
1629
  "checkpoint_release": null,
1630
  "training_epochs": null,
 
1640
  "samples_required": 200,
1641
  "successes": null,
1642
  "success_percent": null,
1643
+ "status": "unavailable: compatible checkpoint not found",
1644
  "execution_budget": 50,
1645
  "checkpoint_release": null,
1646
  "training_epochs": null,
 
1656
  "samples_required": 200,
1657
  "successes": null,
1658
  "success_percent": null,
1659
+ "status": "unavailable: compatible checkpoint not found",
1660
  "execution_budget": 50,
1661
  "checkpoint_release": null,
1662
  "training_epochs": null,
 
1672
  "samples_required": 200,
1673
  "successes": null,
1674
  "success_percent": null,
1675
+ "status": "unavailable: compatible checkpoint not found",
1676
  "execution_budget": 50,
1677
  "checkpoint_release": null,
1678
  "training_epochs": null,
 
1688
  "samples_required": 200,
1689
  "successes": null,
1690
  "success_percent": null,
1691
+ "status": "unavailable: compatible checkpoint not found",
1692
  "execution_budget": 50,
1693
  "checkpoint_release": null,
1694
  "training_epochs": null,
 
1704
  "samples_required": 200,
1705
  "successes": null,
1706
  "success_percent": null,
1707
+ "status": "unavailable: compatible checkpoint not found",
1708
  "execution_budget": 50,
1709
  "checkpoint_release": null,
1710
  "training_epochs": null,
 
1720
  "samples_required": 200,
1721
  "successes": null,
1722
  "success_percent": null,
1723
+ "status": "unavailable: compatible checkpoint not found",
1724
  "execution_budget": 50,
1725
  "checkpoint_release": null,
1726
  "training_epochs": null,
 
1736
  "samples_required": 200,
1737
  "successes": null,
1738
  "success_percent": null,
1739
+ "status": "unavailable: compatible checkpoint not found",
1740
  "execution_budget": 50,
1741
  "checkpoint_release": null,
1742
  "training_epochs": null,
 
1752
  "samples_required": 200,
1753
  "successes": null,
1754
  "success_percent": null,
1755
+ "status": "unavailable: compatible checkpoint not found",
1756
  "execution_budget": 50,
1757
  "checkpoint_release": null,
1758
  "training_epochs": null,
 
1768
  "samples_required": 200,
1769
  "successes": null,
1770
  "success_percent": null,
1771
+ "status": "unavailable: compatible checkpoint not found",
1772
  "execution_budget": 50,
1773
  "checkpoint_release": null,
1774
  "training_epochs": null,
 
1784
  "samples_required": 200,
1785
  "successes": null,
1786
  "success_percent": null,
1787
+ "status": "unavailable: compatible checkpoint not found",
1788
  "execution_budget": 50,
1789
  "checkpoint_release": null,
1790
  "training_epochs": null,
 
1800
  "samples_required": 200,
1801
  "successes": null,
1802
  "success_percent": null,
1803
+ "status": "unavailable: compatible checkpoint not found",
1804
  "execution_budget": 50,
1805
  "checkpoint_release": null,
1806
  "training_epochs": null,
 
1816
  "samples_required": 200,
1817
  "successes": null,
1818
  "success_percent": null,
1819
+ "status": "unavailable: compatible checkpoint not found",
1820
  "execution_budget": 50,
1821
  "checkpoint_release": null,
1822
  "training_epochs": null,
 
1832
  "samples_required": 200,
1833
  "successes": null,
1834
  "success_percent": null,
1835
+ "status": "unavailable: compatible checkpoint not found",
1836
  "execution_budget": 50,
1837
  "checkpoint_release": null,
1838
  "training_epochs": null,
 
1848
  "samples_required": 200,
1849
  "successes": null,
1850
  "success_percent": null,
1851
+ "status": "unavailable: compatible checkpoint not found",
1852
  "execution_budget": 50,
1853
  "checkpoint_release": null,
1854
  "training_epochs": null,
 
1864
  "samples_required": 200,
1865
  "successes": null,
1866
  "success_percent": null,
1867
+ "status": "unavailable: compatible checkpoint not found",
1868
  "execution_budget": 50,
1869
  "checkpoint_release": null,
1870
  "training_epochs": null,
 
1880
  "samples_required": 200,
1881
  "successes": null,
1882
  "success_percent": null,
1883
+ "status": "unavailable: compatible checkpoint not found",
1884
  "execution_budget": 50,
1885
  "checkpoint_release": null,
1886
  "training_epochs": null,
 
1896
  "samples_required": 200,
1897
  "successes": null,
1898
  "success_percent": null,
1899
+ "status": "unavailable: compatible checkpoint not found",
1900
  "execution_budget": 50,
1901
  "checkpoint_release": null,
1902
  "training_epochs": null,
 
1912
  "samples_required": 200,
1913
  "successes": null,
1914
  "success_percent": null,
1915
+ "status": "unavailable: compatible checkpoint not found",
1916
  "execution_budget": 50,
1917
  "checkpoint_release": null,
1918
  "training_epochs": null,
results/planning_success.xlsx CHANGED
Binary files a/results/planning_success.xlsx and b/results/planning_success.xlsx differ
 
results/planning_summary.csv CHANGED
@@ -15,10 +15,10 @@ pldm,tworoom,25,96.0,94.5,95.5,94.0,97.0,5,95.4,1.1937336386313322,95.40 ± 1.19
15
  pldm,tworoom,50,80.0,76.0,81.0,78.5,79.5,5,79.0,1.9039432764659772,79.00 ± 1.90
16
  pldm,tworoom,75,57.5,54.5,52.5,54.5,56.0,5,55.0,1.8708286933869707,55.00 ± 1.87
17
  pldm,tworoom,100,33.5,33.5,33.5,34.5,32.0,5,33.4,0.8944271909999159,33.40 ± 0.89
18
- pldm,cube,25,,,,,,0,,,
19
- pldm,cube,50,,,,,,0,,,
20
- pldm,cube,75,,,,,,0,,,
21
- pldm,cube,100,,,,,,0,,,
22
  pldm,pusht,25,,,,,,0,,,
23
  pldm,pusht,50,,,,,,0,,,
24
  pldm,pusht,75,,,,,,0,,,
 
15
  pldm,tworoom,50,80.0,76.0,81.0,78.5,79.5,5,79.0,1.9039432764659772,79.00 ± 1.90
16
  pldm,tworoom,75,57.5,54.5,52.5,54.5,56.0,5,55.0,1.8708286933869707,55.00 ± 1.87
17
  pldm,tworoom,100,33.5,33.5,33.5,34.5,32.0,5,33.4,0.8944271909999159,33.40 ± 0.89
18
+ pldm,cube,25,55.0,52.0,56.0,56.5,66.0,5,57.1,5.272570530585627,57.10 ± 5.27
19
+ pldm,cube,50,49.5,39.0,41.5,44.5,50.5,5,45.0,4.9749371855331,45.00 ± 4.97
20
+ pldm,cube,75,54.5,58.5,57.0,54.5,51.5,5,55.2,2.6832815729997477,55.20 ± 2.68
21
+ pldm,cube,100,57.5,58.0,51.5,54.0,53.0,5,54.8,2.8416544476765644,54.80 ± 2.84
22
  pldm,pusht,25,,,,,,0,,,
23
  pldm,pusht,50,,,,,,0,,,
24
  pldm,pusht,75,,,,,,0,,,
results/planning_summary.json CHANGED
@@ -227,57 +227,57 @@
227
  "model": "pldm",
228
  "dataset": "cube",
229
  "offset": 25,
230
- "seed_42": null,
231
- "seed_43": null,
232
- "seed_44": null,
233
- "seed_45": null,
234
- "seed_46": null,
235
- "seeds_completed": 0,
236
- "mean_percent": null,
237
- "std_percent": null,
238
- "mean_plus_std": null
239
  },
240
  {
241
  "model": "pldm",
242
  "dataset": "cube",
243
  "offset": 50,
244
- "seed_42": null,
245
- "seed_43": null,
246
- "seed_44": null,
247
- "seed_45": null,
248
- "seed_46": null,
249
- "seeds_completed": 0,
250
- "mean_percent": null,
251
- "std_percent": null,
252
- "mean_plus_std": null
253
  },
254
  {
255
  "model": "pldm",
256
  "dataset": "cube",
257
  "offset": 75,
258
- "seed_42": null,
259
- "seed_43": null,
260
- "seed_44": null,
261
- "seed_45": null,
262
- "seed_46": null,
263
- "seeds_completed": 0,
264
- "mean_percent": null,
265
- "std_percent": null,
266
- "mean_plus_std": null
267
  },
268
  {
269
  "model": "pldm",
270
  "dataset": "cube",
271
  "offset": 100,
272
- "seed_42": null,
273
- "seed_43": null,
274
- "seed_44": null,
275
- "seed_45": null,
276
- "seed_46": null,
277
- "seeds_completed": 0,
278
- "mean_percent": null,
279
- "std_percent": null,
280
- "mean_plus_std": null
281
  },
282
  {
283
  "model": "pldm",
 
227
  "model": "pldm",
228
  "dataset": "cube",
229
  "offset": 25,
230
+ "seed_42": 55.0,
231
+ "seed_43": 52.0,
232
+ "seed_44": 56.0,
233
+ "seed_45": 56.5,
234
+ "seed_46": 66.0,
235
+ "seeds_completed": 5,
236
+ "mean_percent": 57.1,
237
+ "std_percent": 5.272570530585627,
238
+ "mean_plus_std": "57.10 \u00b1 5.27"
239
  },
240
  {
241
  "model": "pldm",
242
  "dataset": "cube",
243
  "offset": 50,
244
+ "seed_42": 49.5,
245
+ "seed_43": 39.0,
246
+ "seed_44": 41.5,
247
+ "seed_45": 44.5,
248
+ "seed_46": 50.5,
249
+ "seeds_completed": 5,
250
+ "mean_percent": 45.0,
251
+ "std_percent": 4.9749371855331,
252
+ "mean_plus_std": "45.00 \u00b1 4.97"
253
  },
254
  {
255
  "model": "pldm",
256
  "dataset": "cube",
257
  "offset": 75,
258
+ "seed_42": 54.5,
259
+ "seed_43": 58.5,
260
+ "seed_44": 57.0,
261
+ "seed_45": 54.5,
262
+ "seed_46": 51.5,
263
+ "seeds_completed": 5,
264
+ "mean_percent": 55.2,
265
+ "std_percent": 2.6832815729997477,
266
+ "mean_plus_std": "55.20 \u00b1 2.68"
267
  },
268
  {
269
  "model": "pldm",
270
  "dataset": "cube",
271
  "offset": 100,
272
+ "seed_42": 57.5,
273
+ "seed_43": 58.0,
274
+ "seed_44": 51.5,
275
+ "seed_45": 54.0,
276
+ "seed_46": 53.0,
277
+ "seeds_completed": 5,
278
+ "mean_percent": 54.8,
279
+ "std_percent": 2.8416544476765644,
280
+ "mean_plus_std": "54.80 \u00b1 2.84"
281
  },
282
  {
283
  "model": "pldm",
results/pldm_cube_100.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 100,
5
+ "seed": 42,
6
+ "successes": [
7
+ true,
8
+ false,
9
+ false,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ false,
16
+ false,
17
+ true,
18
+ true,
19
+ false,
20
+ false,
21
+ false,
22
+ true,
23
+ false,
24
+ true,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ true,
36
+ true,
37
+ true,
38
+ true,
39
+ false,
40
+ true,
41
+ false,
42
+ true,
43
+ false,
44
+ true,
45
+ false,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ true,
51
+ false,
52
+ true,
53
+ false,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ true,
66
+ false,
67
+ true,
68
+ true,
69
+ true,
70
+ false,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ false,
78
+ true,
79
+ false,
80
+ true,
81
+ true,
82
+ false,
83
+ false,
84
+ true,
85
+ true,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ false,
91
+ true,
92
+ false,
93
+ false,
94
+ true,
95
+ true,
96
+ true,
97
+ true,
98
+ false,
99
+ true,
100
+ true,
101
+ false,
102
+ true,
103
+ false,
104
+ false,
105
+ false,
106
+ true,
107
+ false,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ true,
113
+ true,
114
+ false,
115
+ true,
116
+ true,
117
+ false,
118
+ true,
119
+ true,
120
+ true,
121
+ false,
122
+ true,
123
+ true,
124
+ false,
125
+ true,
126
+ true,
127
+ true,
128
+ true,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ false,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ true,
143
+ true,
144
+ false,
145
+ true,
146
+ false,
147
+ true,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ true,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ true,
162
+ false,
163
+ true,
164
+ true,
165
+ true,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ false,
175
+ true,
176
+ false,
177
+ false,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ false,
184
+ true,
185
+ false,
186
+ false,
187
+ true,
188
+ false,
189
+ false,
190
+ true,
191
+ true,
192
+ false,
193
+ false,
194
+ false,
195
+ true,
196
+ false,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ false,
203
+ false,
204
+ true,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 246.83161735534668,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_100_seed43.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 100,
5
+ "seed": 43,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ true,
15
+ true,
16
+ true,
17
+ false,
18
+ true,
19
+ true,
20
+ false,
21
+ true,
22
+ false,
23
+ true,
24
+ false,
25
+ true,
26
+ true,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ true,
38
+ true,
39
+ true,
40
+ true,
41
+ false,
42
+ true,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ false,
48
+ false,
49
+ true,
50
+ true,
51
+ false,
52
+ true,
53
+ true,
54
+ true,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ true,
61
+ false,
62
+ true,
63
+ false,
64
+ false,
65
+ true,
66
+ true,
67
+ true,
68
+ true,
69
+ false,
70
+ true,
71
+ false,
72
+ false,
73
+ true,
74
+ true,
75
+ true,
76
+ true,
77
+ false,
78
+ false,
79
+ true,
80
+ false,
81
+ true,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ true,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ true,
93
+ true,
94
+ false,
95
+ false,
96
+ false,
97
+ false,
98
+ true,
99
+ false,
100
+ true,
101
+ false,
102
+ true,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ false,
110
+ true,
111
+ true,
112
+ true,
113
+ true,
114
+ false,
115
+ false,
116
+ false,
117
+ true,
118
+ true,
119
+ false,
120
+ true,
121
+ false,
122
+ false,
123
+ false,
124
+ true,
125
+ true,
126
+ true,
127
+ true,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ true,
133
+ false,
134
+ false,
135
+ false,
136
+ true,
137
+ false,
138
+ true,
139
+ true,
140
+ true,
141
+ true,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ true,
151
+ false,
152
+ true,
153
+ true,
154
+ true,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ false,
162
+ true,
163
+ true,
164
+ true,
165
+ false,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ false,
175
+ true,
176
+ false,
177
+ true,
178
+ true,
179
+ false,
180
+ true,
181
+ false,
182
+ true,
183
+ false,
184
+ true,
185
+ true,
186
+ false,
187
+ true,
188
+ false,
189
+ true,
190
+ true,
191
+ false,
192
+ true,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ false,
198
+ false,
199
+ true,
200
+ false,
201
+ false,
202
+ false,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 250.5033004283905,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_100_seed44.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 100,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ false,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ false,
18
+ true,
19
+ false,
20
+ false,
21
+ false,
22
+ true,
23
+ true,
24
+ false,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ true,
30
+ false,
31
+ true,
32
+ true,
33
+ false,
34
+ true,
35
+ true,
36
+ true,
37
+ false,
38
+ false,
39
+ false,
40
+ true,
41
+ true,
42
+ true,
43
+ false,
44
+ true,
45
+ true,
46
+ false,
47
+ true,
48
+ true,
49
+ true,
50
+ false,
51
+ true,
52
+ true,
53
+ true,
54
+ true,
55
+ true,
56
+ false,
57
+ false,
58
+ true,
59
+ false,
60
+ true,
61
+ true,
62
+ true,
63
+ true,
64
+ true,
65
+ true,
66
+ false,
67
+ true,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ false,
73
+ false,
74
+ true,
75
+ true,
76
+ true,
77
+ false,
78
+ true,
79
+ false,
80
+ true,
81
+ false,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ false,
91
+ true,
92
+ false,
93
+ true,
94
+ true,
95
+ true,
96
+ false,
97
+ false,
98
+ true,
99
+ false,
100
+ true,
101
+ true,
102
+ true,
103
+ true,
104
+ false,
105
+ false,
106
+ true,
107
+ true,
108
+ false,
109
+ true,
110
+ false,
111
+ true,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ true,
117
+ false,
118
+ true,
119
+ false,
120
+ true,
121
+ false,
122
+ true,
123
+ false,
124
+ false,
125
+ true,
126
+ true,
127
+ false,
128
+ false,
129
+ false,
130
+ false,
131
+ false,
132
+ true,
133
+ false,
134
+ true,
135
+ false,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ true,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ true,
149
+ false,
150
+ false,
151
+ false,
152
+ true,
153
+ false,
154
+ true,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ false,
161
+ true,
162
+ false,
163
+ true,
164
+ true,
165
+ false,
166
+ true,
167
+ false,
168
+ true,
169
+ false,
170
+ true,
171
+ false,
172
+ true,
173
+ true,
174
+ true,
175
+ false,
176
+ true,
177
+ false,
178
+ false,
179
+ true,
180
+ true,
181
+ false,
182
+ true,
183
+ false,
184
+ false,
185
+ true,
186
+ true,
187
+ false,
188
+ false,
189
+ false,
190
+ true,
191
+ true,
192
+ true,
193
+ true,
194
+ true,
195
+ false,
196
+ true,
197
+ false,
198
+ true,
199
+ false,
200
+ false,
201
+ true,
202
+ false,
203
+ false,
204
+ true,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 256.6976988315582,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_100_seed45.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 100,
5
+ "seed": 45,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ true,
13
+ false,
14
+ true,
15
+ true,
16
+ false,
17
+ false,
18
+ true,
19
+ false,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ true,
26
+ false,
27
+ true,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ false,
37
+ true,
38
+ false,
39
+ false,
40
+ false,
41
+ true,
42
+ true,
43
+ true,
44
+ false,
45
+ false,
46
+ false,
47
+ true,
48
+ true,
49
+ false,
50
+ false,
51
+ true,
52
+ false,
53
+ true,
54
+ true,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ true,
62
+ true,
63
+ true,
64
+ true,
65
+ true,
66
+ false,
67
+ true,
68
+ false,
69
+ false,
70
+ true,
71
+ false,
72
+ true,
73
+ true,
74
+ false,
75
+ true,
76
+ false,
77
+ true,
78
+ true,
79
+ true,
80
+ true,
81
+ false,
82
+ true,
83
+ false,
84
+ false,
85
+ true,
86
+ false,
87
+ true,
88
+ false,
89
+ false,
90
+ true,
91
+ true,
92
+ true,
93
+ true,
94
+ false,
95
+ false,
96
+ true,
97
+ true,
98
+ true,
99
+ true,
100
+ false,
101
+ false,
102
+ false,
103
+ true,
104
+ true,
105
+ true,
106
+ false,
107
+ false,
108
+ true,
109
+ true,
110
+ false,
111
+ false,
112
+ false,
113
+ false,
114
+ true,
115
+ false,
116
+ true,
117
+ false,
118
+ false,
119
+ true,
120
+ false,
121
+ true,
122
+ false,
123
+ false,
124
+ true,
125
+ false,
126
+ true,
127
+ false,
128
+ true,
129
+ false,
130
+ true,
131
+ false,
132
+ true,
133
+ true,
134
+ false,
135
+ false,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ true,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ false,
148
+ true,
149
+ false,
150
+ true,
151
+ false,
152
+ true,
153
+ true,
154
+ true,
155
+ true,
156
+ false,
157
+ false,
158
+ false,
159
+ true,
160
+ false,
161
+ false,
162
+ true,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ false,
171
+ true,
172
+ false,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ false,
179
+ false,
180
+ false,
181
+ true,
182
+ false,
183
+ false,
184
+ false,
185
+ true,
186
+ false,
187
+ false,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ false,
195
+ true,
196
+ true,
197
+ true,
198
+ true,
199
+ false,
200
+ true,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 254.08669662475586,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_100_seed46.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 100,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ false,
13
+ true,
14
+ true,
15
+ false,
16
+ true,
17
+ false,
18
+ true,
19
+ false,
20
+ true,
21
+ true,
22
+ false,
23
+ false,
24
+ false,
25
+ true,
26
+ false,
27
+ true,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ false,
36
+ true,
37
+ true,
38
+ true,
39
+ false,
40
+ true,
41
+ true,
42
+ false,
43
+ true,
44
+ true,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ true,
51
+ false,
52
+ false,
53
+ false,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ false,
59
+ false,
60
+ true,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ true,
67
+ true,
68
+ false,
69
+ false,
70
+ true,
71
+ false,
72
+ true,
73
+ false,
74
+ true,
75
+ false,
76
+ true,
77
+ true,
78
+ false,
79
+ true,
80
+ false,
81
+ true,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ true,
88
+ true,
89
+ true,
90
+ false,
91
+ false,
92
+ false,
93
+ true,
94
+ true,
95
+ false,
96
+ false,
97
+ false,
98
+ true,
99
+ true,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ true,
105
+ false,
106
+ true,
107
+ true,
108
+ false,
109
+ false,
110
+ true,
111
+ true,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ false,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ true,
124
+ true,
125
+ true,
126
+ false,
127
+ true,
128
+ false,
129
+ true,
130
+ false,
131
+ true,
132
+ false,
133
+ false,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ true,
139
+ false,
140
+ true,
141
+ true,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ false,
148
+ false,
149
+ true,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ true,
158
+ true,
159
+ false,
160
+ true,
161
+ false,
162
+ true,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ true,
168
+ false,
169
+ true,
170
+ false,
171
+ true,
172
+ false,
173
+ true,
174
+ true,
175
+ false,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ false,
182
+ true,
183
+ true,
184
+ false,
185
+ true,
186
+ false,
187
+ true,
188
+ false,
189
+ true,
190
+ false,
191
+ false,
192
+ true,
193
+ false,
194
+ true,
195
+ true,
196
+ true,
197
+ false,
198
+ true,
199
+ false,
200
+ false,
201
+ true,
202
+ false,
203
+ true,
204
+ true,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 261.3357481956482,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_25.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 25,
5
+ "seed": 42,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ true,
11
+ false,
12
+ true,
13
+ true,
14
+ false,
15
+ false,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ false,
24
+ true,
25
+ false,
26
+ true,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ true,
34
+ false,
35
+ true,
36
+ true,
37
+ true,
38
+ false,
39
+ false,
40
+ true,
41
+ true,
42
+ false,
43
+ false,
44
+ true,
45
+ true,
46
+ false,
47
+ true,
48
+ false,
49
+ false,
50
+ false,
51
+ true,
52
+ true,
53
+ true,
54
+ false,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ true,
61
+ true,
62
+ false,
63
+ true,
64
+ true,
65
+ false,
66
+ true,
67
+ true,
68
+ true,
69
+ true,
70
+ false,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ false,
77
+ true,
78
+ false,
79
+ false,
80
+ false,
81
+ true,
82
+ true,
83
+ true,
84
+ false,
85
+ true,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ true,
91
+ true,
92
+ true,
93
+ true,
94
+ false,
95
+ true,
96
+ true,
97
+ false,
98
+ true,
99
+ false,
100
+ false,
101
+ true,
102
+ true,
103
+ false,
104
+ false,
105
+ true,
106
+ true,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ false,
112
+ false,
113
+ false,
114
+ false,
115
+ true,
116
+ false,
117
+ false,
118
+ true,
119
+ false,
120
+ false,
121
+ true,
122
+ false,
123
+ false,
124
+ false,
125
+ true,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ false,
137
+ true,
138
+ false,
139
+ false,
140
+ false,
141
+ true,
142
+ true,
143
+ false,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ true,
152
+ true,
153
+ false,
154
+ false,
155
+ true,
156
+ true,
157
+ false,
158
+ false,
159
+ false,
160
+ true,
161
+ true,
162
+ true,
163
+ false,
164
+ false,
165
+ true,
166
+ true,
167
+ true,
168
+ true,
169
+ true,
170
+ true,
171
+ false,
172
+ true,
173
+ true,
174
+ true,
175
+ true,
176
+ true,
177
+ false,
178
+ false,
179
+ false,
180
+ false,
181
+ true,
182
+ false,
183
+ true,
184
+ false,
185
+ true,
186
+ true,
187
+ false,
188
+ true,
189
+ false,
190
+ true,
191
+ true,
192
+ false,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ false,
202
+ false,
203
+ false,
204
+ true,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 255.5752260684967,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_25_seed43.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 25,
5
+ "seed": 43,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ false,
13
+ true,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ true,
19
+ false,
20
+ true,
21
+ false,
22
+ false,
23
+ true,
24
+ true,
25
+ false,
26
+ true,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ false,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ false,
40
+ true,
41
+ false,
42
+ true,
43
+ true,
44
+ true,
45
+ false,
46
+ false,
47
+ false,
48
+ true,
49
+ true,
50
+ true,
51
+ true,
52
+ true,
53
+ false,
54
+ false,
55
+ true,
56
+ false,
57
+ false,
58
+ true,
59
+ false,
60
+ true,
61
+ true,
62
+ true,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ false,
69
+ true,
70
+ true,
71
+ true,
72
+ true,
73
+ true,
74
+ true,
75
+ true,
76
+ true,
77
+ true,
78
+ false,
79
+ true,
80
+ false,
81
+ false,
82
+ true,
83
+ false,
84
+ true,
85
+ false,
86
+ true,
87
+ true,
88
+ true,
89
+ true,
90
+ true,
91
+ false,
92
+ false,
93
+ true,
94
+ false,
95
+ true,
96
+ true,
97
+ false,
98
+ false,
99
+ true,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ false,
110
+ true,
111
+ true,
112
+ false,
113
+ true,
114
+ false,
115
+ false,
116
+ true,
117
+ true,
118
+ true,
119
+ true,
120
+ false,
121
+ true,
122
+ true,
123
+ false,
124
+ true,
125
+ true,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ false,
133
+ true,
134
+ true,
135
+ false,
136
+ false,
137
+ true,
138
+ true,
139
+ true,
140
+ true,
141
+ false,
142
+ true,
143
+ true,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ true,
149
+ false,
150
+ true,
151
+ false,
152
+ false,
153
+ true,
154
+ false,
155
+ false,
156
+ false,
157
+ true,
158
+ true,
159
+ true,
160
+ false,
161
+ false,
162
+ false,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ false,
171
+ false,
172
+ true,
173
+ true,
174
+ true,
175
+ false,
176
+ false,
177
+ false,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ false,
183
+ true,
184
+ true,
185
+ false,
186
+ false,
187
+ false,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ true,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ false,
198
+ true,
199
+ false,
200
+ false,
201
+ false,
202
+ true,
203
+ true,
204
+ false,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 256.3706622123718,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_25_seed44.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 25,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ false,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ false,
21
+ false,
22
+ true,
23
+ false,
24
+ true,
25
+ true,
26
+ false,
27
+ true,
28
+ false,
29
+ false,
30
+ false,
31
+ true,
32
+ true,
33
+ false,
34
+ true,
35
+ true,
36
+ false,
37
+ false,
38
+ false,
39
+ true,
40
+ true,
41
+ true,
42
+ false,
43
+ true,
44
+ true,
45
+ false,
46
+ false,
47
+ false,
48
+ true,
49
+ false,
50
+ false,
51
+ false,
52
+ true,
53
+ true,
54
+ true,
55
+ false,
56
+ false,
57
+ false,
58
+ true,
59
+ true,
60
+ true,
61
+ true,
62
+ false,
63
+ false,
64
+ true,
65
+ true,
66
+ false,
67
+ false,
68
+ true,
69
+ false,
70
+ false,
71
+ false,
72
+ false,
73
+ true,
74
+ true,
75
+ false,
76
+ true,
77
+ true,
78
+ true,
79
+ false,
80
+ false,
81
+ true,
82
+ false,
83
+ false,
84
+ true,
85
+ true,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ true,
91
+ true,
92
+ true,
93
+ false,
94
+ false,
95
+ true,
96
+ true,
97
+ false,
98
+ true,
99
+ true,
100
+ false,
101
+ false,
102
+ true,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ true,
110
+ false,
111
+ false,
112
+ true,
113
+ true,
114
+ false,
115
+ true,
116
+ true,
117
+ true,
118
+ true,
119
+ true,
120
+ true,
121
+ false,
122
+ true,
123
+ true,
124
+ true,
125
+ false,
126
+ false,
127
+ true,
128
+ true,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ true,
135
+ true,
136
+ true,
137
+ true,
138
+ true,
139
+ false,
140
+ true,
141
+ false,
142
+ false,
143
+ true,
144
+ true,
145
+ false,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ false,
151
+ true,
152
+ true,
153
+ false,
154
+ true,
155
+ false,
156
+ true,
157
+ true,
158
+ false,
159
+ true,
160
+ false,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ true,
167
+ false,
168
+ false,
169
+ true,
170
+ false,
171
+ true,
172
+ true,
173
+ false,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ false,
180
+ false,
181
+ false,
182
+ true,
183
+ true,
184
+ true,
185
+ false,
186
+ false,
187
+ true,
188
+ false,
189
+ false,
190
+ false,
191
+ true,
192
+ true,
193
+ false,
194
+ false,
195
+ false,
196
+ true,
197
+ true,
198
+ false,
199
+ true,
200
+ false,
201
+ true,
202
+ true,
203
+ true,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 252.09164381027222,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_25_seed45.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 25,
5
+ "seed": 45,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ false,
12
+ true,
13
+ true,
14
+ true,
15
+ false,
16
+ true,
17
+ false,
18
+ false,
19
+ false,
20
+ false,
21
+ true,
22
+ true,
23
+ false,
24
+ true,
25
+ true,
26
+ false,
27
+ false,
28
+ false,
29
+ false,
30
+ true,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ false,
36
+ false,
37
+ true,
38
+ true,
39
+ false,
40
+ false,
41
+ true,
42
+ false,
43
+ true,
44
+ true,
45
+ false,
46
+ true,
47
+ false,
48
+ false,
49
+ false,
50
+ true,
51
+ true,
52
+ false,
53
+ true,
54
+ true,
55
+ true,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ true,
61
+ true,
62
+ true,
63
+ true,
64
+ false,
65
+ true,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ true,
71
+ false,
72
+ false,
73
+ true,
74
+ true,
75
+ true,
76
+ true,
77
+ false,
78
+ true,
79
+ false,
80
+ true,
81
+ false,
82
+ false,
83
+ false,
84
+ true,
85
+ false,
86
+ false,
87
+ true,
88
+ true,
89
+ false,
90
+ true,
91
+ false,
92
+ true,
93
+ true,
94
+ true,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ true,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ true,
105
+ true,
106
+ false,
107
+ true,
108
+ true,
109
+ true,
110
+ false,
111
+ false,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ false,
117
+ true,
118
+ true,
119
+ true,
120
+ false,
121
+ true,
122
+ false,
123
+ false,
124
+ false,
125
+ true,
126
+ false,
127
+ false,
128
+ false,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ false,
135
+ false,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ true,
141
+ true,
142
+ true,
143
+ false,
144
+ true,
145
+ false,
146
+ false,
147
+ false,
148
+ true,
149
+ false,
150
+ true,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ true,
162
+ false,
163
+ false,
164
+ true,
165
+ true,
166
+ false,
167
+ true,
168
+ true,
169
+ true,
170
+ true,
171
+ false,
172
+ true,
173
+ false,
174
+ true,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ true,
184
+ true,
185
+ true,
186
+ false,
187
+ true,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ true,
195
+ true,
196
+ false,
197
+ false,
198
+ true,
199
+ true,
200
+ false,
201
+ true,
202
+ false,
203
+ true,
204
+ true,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 251.9367880821228,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_25_seed46.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 25,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ true,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ true,
26
+ false,
27
+ true,
28
+ true,
29
+ false,
30
+ true,
31
+ true,
32
+ true,
33
+ true,
34
+ false,
35
+ true,
36
+ false,
37
+ true,
38
+ true,
39
+ false,
40
+ false,
41
+ true,
42
+ false,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ false,
51
+ false,
52
+ true,
53
+ false,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ false,
61
+ true,
62
+ false,
63
+ true,
64
+ true,
65
+ true,
66
+ false,
67
+ true,
68
+ true,
69
+ false,
70
+ true,
71
+ true,
72
+ true,
73
+ true,
74
+ false,
75
+ false,
76
+ true,
77
+ true,
78
+ true,
79
+ true,
80
+ true,
81
+ true,
82
+ false,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ true,
88
+ false,
89
+ true,
90
+ false,
91
+ false,
92
+ false,
93
+ true,
94
+ true,
95
+ true,
96
+ true,
97
+ false,
98
+ true,
99
+ false,
100
+ true,
101
+ false,
102
+ false,
103
+ true,
104
+ true,
105
+ false,
106
+ true,
107
+ true,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ true,
117
+ true,
118
+ false,
119
+ true,
120
+ true,
121
+ true,
122
+ true,
123
+ true,
124
+ false,
125
+ true,
126
+ false,
127
+ true,
128
+ false,
129
+ true,
130
+ false,
131
+ true,
132
+ false,
133
+ true,
134
+ true,
135
+ false,
136
+ false,
137
+ false,
138
+ false,
139
+ true,
140
+ true,
141
+ false,
142
+ true,
143
+ true,
144
+ false,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ true,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ true,
158
+ false,
159
+ true,
160
+ true,
161
+ false,
162
+ true,
163
+ true,
164
+ false,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ false,
171
+ true,
172
+ true,
173
+ false,
174
+ true,
175
+ false,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ false,
182
+ false,
183
+ false,
184
+ true,
185
+ true,
186
+ true,
187
+ true,
188
+ false,
189
+ true,
190
+ true,
191
+ true,
192
+ true,
193
+ true,
194
+ true,
195
+ false,
196
+ true,
197
+ true,
198
+ true,
199
+ true,
200
+ false,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 250.82570028305054,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_50.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 50,
5
+ "seed": 42,
6
+ "successes": [
7
+ false,
8
+ false,
9
+ false,
10
+ true,
11
+ false,
12
+ true,
13
+ false,
14
+ false,
15
+ false,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ false,
24
+ false,
25
+ false,
26
+ true,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ false,
35
+ true,
36
+ true,
37
+ true,
38
+ true,
39
+ false,
40
+ true,
41
+ false,
42
+ false,
43
+ false,
44
+ true,
45
+ true,
46
+ true,
47
+ false,
48
+ true,
49
+ false,
50
+ false,
51
+ true,
52
+ false,
53
+ true,
54
+ false,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ true,
61
+ true,
62
+ false,
63
+ true,
64
+ true,
65
+ false,
66
+ true,
67
+ false,
68
+ true,
69
+ false,
70
+ false,
71
+ true,
72
+ true,
73
+ false,
74
+ false,
75
+ false,
76
+ true,
77
+ false,
78
+ true,
79
+ false,
80
+ true,
81
+ false,
82
+ true,
83
+ false,
84
+ false,
85
+ true,
86
+ true,
87
+ false,
88
+ true,
89
+ true,
90
+ false,
91
+ true,
92
+ false,
93
+ false,
94
+ false,
95
+ false,
96
+ true,
97
+ false,
98
+ true,
99
+ false,
100
+ true,
101
+ true,
102
+ true,
103
+ false,
104
+ false,
105
+ true,
106
+ true,
107
+ true,
108
+ false,
109
+ true,
110
+ false,
111
+ false,
112
+ true,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ true,
118
+ false,
119
+ false,
120
+ false,
121
+ true,
122
+ false,
123
+ true,
124
+ false,
125
+ true,
126
+ false,
127
+ true,
128
+ true,
129
+ false,
130
+ true,
131
+ false,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ false,
137
+ true,
138
+ false,
139
+ true,
140
+ true,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ true,
149
+ false,
150
+ true,
151
+ true,
152
+ true,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ false,
158
+ true,
159
+ false,
160
+ true,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ true,
166
+ true,
167
+ true,
168
+ false,
169
+ true,
170
+ true,
171
+ false,
172
+ false,
173
+ false,
174
+ true,
175
+ false,
176
+ true,
177
+ false,
178
+ true,
179
+ true,
180
+ false,
181
+ false,
182
+ true,
183
+ true,
184
+ true,
185
+ true,
186
+ true,
187
+ true,
188
+ true,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ true,
196
+ false,
197
+ true,
198
+ true,
199
+ true,
200
+ true,
201
+ false,
202
+ true,
203
+ false,
204
+ true,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 257.31775760650635,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_50_seed43.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 50,
5
+ "seed": 43,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ false,
10
+ false,
11
+ true,
12
+ false,
13
+ false,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ false,
26
+ true,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ false,
37
+ false,
38
+ true,
39
+ true,
40
+ false,
41
+ false,
42
+ true,
43
+ true,
44
+ true,
45
+ false,
46
+ true,
47
+ false,
48
+ true,
49
+ true,
50
+ true,
51
+ true,
52
+ true,
53
+ false,
54
+ false,
55
+ true,
56
+ false,
57
+ false,
58
+ true,
59
+ false,
60
+ true,
61
+ true,
62
+ false,
63
+ false,
64
+ true,
65
+ false,
66
+ true,
67
+ true,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ true,
73
+ false,
74
+ true,
75
+ true,
76
+ true,
77
+ true,
78
+ false,
79
+ false,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ false,
86
+ true,
87
+ true,
88
+ false,
89
+ true,
90
+ false,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ true,
96
+ false,
97
+ true,
98
+ true,
99
+ false,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ false,
108
+ true,
109
+ false,
110
+ true,
111
+ false,
112
+ false,
113
+ true,
114
+ false,
115
+ true,
116
+ true,
117
+ false,
118
+ true,
119
+ true,
120
+ true,
121
+ true,
122
+ true,
123
+ false,
124
+ true,
125
+ false,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ false,
132
+ false,
133
+ true,
134
+ false,
135
+ false,
136
+ false,
137
+ true,
138
+ true,
139
+ false,
140
+ true,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ true,
146
+ false,
147
+ false,
148
+ false,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ true,
158
+ false,
159
+ false,
160
+ false,
161
+ false,
162
+ true,
163
+ false,
164
+ false,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ false,
171
+ false,
172
+ false,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ false,
178
+ true,
179
+ true,
180
+ false,
181
+ false,
182
+ true,
183
+ false,
184
+ false,
185
+ false,
186
+ true,
187
+ true,
188
+ true,
189
+ false,
190
+ false,
191
+ false,
192
+ true,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ false,
198
+ false,
199
+ true,
200
+ false,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 266.6438059806824,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_50_seed44.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 50,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ false,
10
+ true,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ false,
21
+ false,
22
+ true,
23
+ false,
24
+ false,
25
+ true,
26
+ false,
27
+ true,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ false,
38
+ false,
39
+ true,
40
+ true,
41
+ false,
42
+ false,
43
+ false,
44
+ false,
45
+ false,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ true,
52
+ false,
53
+ false,
54
+ true,
55
+ false,
56
+ false,
57
+ false,
58
+ true,
59
+ true,
60
+ true,
61
+ true,
62
+ false,
63
+ false,
64
+ false,
65
+ false,
66
+ false,
67
+ false,
68
+ false,
69
+ true,
70
+ false,
71
+ false,
72
+ false,
73
+ true,
74
+ true,
75
+ false,
76
+ true,
77
+ true,
78
+ false,
79
+ false,
80
+ false,
81
+ true,
82
+ true,
83
+ false,
84
+ true,
85
+ true,
86
+ false,
87
+ false,
88
+ true,
89
+ true,
90
+ true,
91
+ false,
92
+ false,
93
+ false,
94
+ false,
95
+ true,
96
+ true,
97
+ false,
98
+ true,
99
+ false,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ false,
109
+ false,
110
+ false,
111
+ false,
112
+ false,
113
+ true,
114
+ false,
115
+ true,
116
+ true,
117
+ true,
118
+ true,
119
+ false,
120
+ true,
121
+ false,
122
+ true,
123
+ true,
124
+ false,
125
+ false,
126
+ false,
127
+ false,
128
+ true,
129
+ false,
130
+ false,
131
+ false,
132
+ false,
133
+ false,
134
+ true,
135
+ false,
136
+ true,
137
+ true,
138
+ true,
139
+ true,
140
+ false,
141
+ false,
142
+ false,
143
+ true,
144
+ false,
145
+ false,
146
+ true,
147
+ false,
148
+ true,
149
+ false,
150
+ false,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ true,
157
+ true,
158
+ false,
159
+ false,
160
+ false,
161
+ true,
162
+ false,
163
+ false,
164
+ false,
165
+ false,
166
+ true,
167
+ false,
168
+ false,
169
+ false,
170
+ true,
171
+ false,
172
+ true,
173
+ false,
174
+ false,
175
+ true,
176
+ true,
177
+ true,
178
+ false,
179
+ false,
180
+ true,
181
+ false,
182
+ false,
183
+ true,
184
+ false,
185
+ false,
186
+ false,
187
+ true,
188
+ true,
189
+ false,
190
+ true,
191
+ false,
192
+ true,
193
+ true,
194
+ true,
195
+ true,
196
+ false,
197
+ true,
198
+ false,
199
+ false,
200
+ true,
201
+ true,
202
+ false,
203
+ true,
204
+ true,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 266.58846044540405,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_50_seed45.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 50,
5
+ "seed": 45,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ false,
10
+ false,
11
+ false,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ false,
19
+ false,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ false,
26
+ false,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ true,
34
+ true,
35
+ false,
36
+ false,
37
+ true,
38
+ false,
39
+ false,
40
+ false,
41
+ false,
42
+ true,
43
+ false,
44
+ false,
45
+ true,
46
+ false,
47
+ false,
48
+ false,
49
+ false,
50
+ true,
51
+ true,
52
+ true,
53
+ false,
54
+ false,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ true,
60
+ true,
61
+ true,
62
+ true,
63
+ false,
64
+ true,
65
+ false,
66
+ false,
67
+ false,
68
+ true,
69
+ true,
70
+ false,
71
+ false,
72
+ true,
73
+ false,
74
+ true,
75
+ true,
76
+ false,
77
+ true,
78
+ false,
79
+ true,
80
+ false,
81
+ false,
82
+ false,
83
+ false,
84
+ false,
85
+ true,
86
+ true,
87
+ true,
88
+ false,
89
+ true,
90
+ true,
91
+ false,
92
+ true,
93
+ true,
94
+ false,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ false,
100
+ true,
101
+ true,
102
+ false,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ false,
108
+ false,
109
+ false,
110
+ false,
111
+ true,
112
+ false,
113
+ true,
114
+ true,
115
+ false,
116
+ true,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ false,
122
+ true,
123
+ false,
124
+ true,
125
+ true,
126
+ true,
127
+ false,
128
+ true,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ false,
135
+ true,
136
+ true,
137
+ false,
138
+ false,
139
+ true,
140
+ true,
141
+ true,
142
+ false,
143
+ true,
144
+ false,
145
+ false,
146
+ false,
147
+ false,
148
+ false,
149
+ true,
150
+ true,
151
+ false,
152
+ false,
153
+ false,
154
+ false,
155
+ false,
156
+ false,
157
+ false,
158
+ false,
159
+ true,
160
+ true,
161
+ true,
162
+ true,
163
+ true,
164
+ true,
165
+ true,
166
+ false,
167
+ false,
168
+ false,
169
+ false,
170
+ false,
171
+ false,
172
+ true,
173
+ false,
174
+ false,
175
+ false,
176
+ true,
177
+ true,
178
+ false,
179
+ true,
180
+ true,
181
+ false,
182
+ false,
183
+ true,
184
+ true,
185
+ false,
186
+ false,
187
+ true,
188
+ true,
189
+ false,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ false,
198
+ true,
199
+ true,
200
+ false,
201
+ true,
202
+ false,
203
+ false,
204
+ true,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 275.976389169693,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_50_seed46.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 50,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ true,
11
+ true,
12
+ true,
13
+ false,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ true,
19
+ true,
20
+ false,
21
+ true,
22
+ true,
23
+ true,
24
+ false,
25
+ true,
26
+ false,
27
+ true,
28
+ true,
29
+ true,
30
+ false,
31
+ true,
32
+ false,
33
+ true,
34
+ false,
35
+ false,
36
+ false,
37
+ true,
38
+ true,
39
+ false,
40
+ false,
41
+ false,
42
+ false,
43
+ true,
44
+ true,
45
+ false,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ false,
51
+ false,
52
+ true,
53
+ false,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ true,
60
+ false,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ true,
69
+ false,
70
+ true,
71
+ true,
72
+ true,
73
+ true,
74
+ false,
75
+ false,
76
+ true,
77
+ true,
78
+ false,
79
+ true,
80
+ true,
81
+ true,
82
+ true,
83
+ true,
84
+ false,
85
+ false,
86
+ false,
87
+ true,
88
+ false,
89
+ false,
90
+ false,
91
+ false,
92
+ false,
93
+ true,
94
+ true,
95
+ false,
96
+ false,
97
+ false,
98
+ true,
99
+ false,
100
+ false,
101
+ true,
102
+ true,
103
+ true,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ false,
109
+ true,
110
+ true,
111
+ false,
112
+ false,
113
+ false,
114
+ true,
115
+ false,
116
+ true,
117
+ false,
118
+ false,
119
+ true,
120
+ false,
121
+ true,
122
+ false,
123
+ true,
124
+ false,
125
+ true,
126
+ false,
127
+ true,
128
+ false,
129
+ false,
130
+ true,
131
+ false,
132
+ false,
133
+ true,
134
+ true,
135
+ true,
136
+ false,
137
+ false,
138
+ false,
139
+ true,
140
+ true,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ false,
151
+ false,
152
+ true,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ false,
158
+ true,
159
+ true,
160
+ false,
161
+ true,
162
+ true,
163
+ true,
164
+ false,
165
+ true,
166
+ false,
167
+ false,
168
+ true,
169
+ true,
170
+ false,
171
+ true,
172
+ true,
173
+ false,
174
+ false,
175
+ false,
176
+ false,
177
+ true,
178
+ true,
179
+ false,
180
+ false,
181
+ false,
182
+ false,
183
+ true,
184
+ false,
185
+ true,
186
+ true,
187
+ true,
188
+ false,
189
+ true,
190
+ false,
191
+ true,
192
+ false,
193
+ true,
194
+ false,
195
+ false,
196
+ true,
197
+ true,
198
+ false,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ false,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 258.9741187095642,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_75.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 75,
5
+ "seed": 42,
6
+ "successes": [
7
+ true,
8
+ false,
9
+ false,
10
+ true,
11
+ true,
12
+ true,
13
+ false,
14
+ false,
15
+ false,
16
+ false,
17
+ true,
18
+ true,
19
+ false,
20
+ false,
21
+ true,
22
+ true,
23
+ true,
24
+ true,
25
+ false,
26
+ true,
27
+ false,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ true,
34
+ false,
35
+ true,
36
+ true,
37
+ true,
38
+ true,
39
+ false,
40
+ true,
41
+ true,
42
+ true,
43
+ false,
44
+ true,
45
+ false,
46
+ true,
47
+ true,
48
+ false,
49
+ false,
50
+ true,
51
+ true,
52
+ true,
53
+ true,
54
+ false,
55
+ false,
56
+ true,
57
+ false,
58
+ true,
59
+ false,
60
+ false,
61
+ false,
62
+ false,
63
+ true,
64
+ true,
65
+ true,
66
+ false,
67
+ true,
68
+ true,
69
+ true,
70
+ false,
71
+ true,
72
+ false,
73
+ false,
74
+ false,
75
+ false,
76
+ true,
77
+ true,
78
+ true,
79
+ false,
80
+ true,
81
+ false,
82
+ false,
83
+ true,
84
+ false,
85
+ true,
86
+ true,
87
+ false,
88
+ false,
89
+ true,
90
+ true,
91
+ false,
92
+ false,
93
+ false,
94
+ true,
95
+ true,
96
+ true,
97
+ true,
98
+ true,
99
+ false,
100
+ true,
101
+ false,
102
+ false,
103
+ false,
104
+ false,
105
+ false,
106
+ true,
107
+ false,
108
+ true,
109
+ true,
110
+ true,
111
+ true,
112
+ true,
113
+ false,
114
+ false,
115
+ false,
116
+ false,
117
+ false,
118
+ true,
119
+ false,
120
+ false,
121
+ false,
122
+ false,
123
+ true,
124
+ false,
125
+ true,
126
+ false,
127
+ true,
128
+ true,
129
+ true,
130
+ false,
131
+ true,
132
+ true,
133
+ true,
134
+ true,
135
+ true,
136
+ false,
137
+ true,
138
+ false,
139
+ true,
140
+ true,
141
+ true,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ true,
152
+ true,
153
+ true,
154
+ true,
155
+ true,
156
+ true,
157
+ false,
158
+ false,
159
+ true,
160
+ true,
161
+ false,
162
+ false,
163
+ true,
164
+ false,
165
+ true,
166
+ false,
167
+ true,
168
+ false,
169
+ true,
170
+ true,
171
+ true,
172
+ true,
173
+ true,
174
+ false,
175
+ true,
176
+ false,
177
+ false,
178
+ false,
179
+ true,
180
+ false,
181
+ true,
182
+ false,
183
+ false,
184
+ true,
185
+ false,
186
+ true,
187
+ true,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ false,
194
+ false,
195
+ true,
196
+ false,
197
+ false,
198
+ false,
199
+ true,
200
+ true,
201
+ false,
202
+ true,
203
+ true,
204
+ true,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 252.26482963562012,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_75_seed43.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 75,
5
+ "seed": 43,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ false,
12
+ false,
13
+ false,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ false,
19
+ true,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ false,
25
+ true,
26
+ true,
27
+ false,
28
+ false,
29
+ false,
30
+ false,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ true,
38
+ true,
39
+ true,
40
+ true,
41
+ false,
42
+ true,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ true,
48
+ false,
49
+ true,
50
+ true,
51
+ false,
52
+ true,
53
+ true,
54
+ true,
55
+ true,
56
+ true,
57
+ false,
58
+ true,
59
+ true,
60
+ true,
61
+ false,
62
+ true,
63
+ false,
64
+ true,
65
+ false,
66
+ true,
67
+ true,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ true,
73
+ true,
74
+ true,
75
+ true,
76
+ true,
77
+ true,
78
+ true,
79
+ false,
80
+ false,
81
+ true,
82
+ false,
83
+ true,
84
+ false,
85
+ true,
86
+ true,
87
+ true,
88
+ false,
89
+ true,
90
+ true,
91
+ false,
92
+ true,
93
+ false,
94
+ false,
95
+ true,
96
+ false,
97
+ false,
98
+ true,
99
+ false,
100
+ false,
101
+ false,
102
+ true,
103
+ true,
104
+ false,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ false,
110
+ true,
111
+ false,
112
+ true,
113
+ true,
114
+ false,
115
+ false,
116
+ false,
117
+ true,
118
+ true,
119
+ true,
120
+ true,
121
+ false,
122
+ true,
123
+ true,
124
+ true,
125
+ false,
126
+ true,
127
+ true,
128
+ true,
129
+ true,
130
+ false,
131
+ false,
132
+ true,
133
+ true,
134
+ false,
135
+ false,
136
+ true,
137
+ true,
138
+ true,
139
+ true,
140
+ true,
141
+ false,
142
+ false,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ true,
148
+ true,
149
+ true,
150
+ false,
151
+ true,
152
+ false,
153
+ true,
154
+ false,
155
+ true,
156
+ true,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ false,
162
+ true,
163
+ false,
164
+ true,
165
+ false,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ false,
171
+ true,
172
+ true,
173
+ true,
174
+ false,
175
+ true,
176
+ true,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ false,
182
+ true,
183
+ false,
184
+ false,
185
+ false,
186
+ true,
187
+ true,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ true,
193
+ false,
194
+ true,
195
+ false,
196
+ false,
197
+ true,
198
+ false,
199
+ false,
200
+ true,
201
+ true,
202
+ true,
203
+ true,
204
+ false,
205
+ false,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 251.66196060180664,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_75_seed44.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 75,
5
+ "seed": 44,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ false,
13
+ true,
14
+ false,
15
+ true,
16
+ true,
17
+ true,
18
+ true,
19
+ true,
20
+ true,
21
+ false,
22
+ true,
23
+ true,
24
+ true,
25
+ true,
26
+ false,
27
+ false,
28
+ false,
29
+ true,
30
+ false,
31
+ true,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ true,
37
+ false,
38
+ false,
39
+ true,
40
+ true,
41
+ true,
42
+ true,
43
+ false,
44
+ true,
45
+ false,
46
+ false,
47
+ true,
48
+ true,
49
+ false,
50
+ false,
51
+ true,
52
+ false,
53
+ false,
54
+ true,
55
+ false,
56
+ false,
57
+ false,
58
+ true,
59
+ true,
60
+ true,
61
+ true,
62
+ false,
63
+ false,
64
+ false,
65
+ true,
66
+ true,
67
+ false,
68
+ false,
69
+ false,
70
+ true,
71
+ false,
72
+ true,
73
+ false,
74
+ true,
75
+ true,
76
+ true,
77
+ true,
78
+ false,
79
+ true,
80
+ false,
81
+ false,
82
+ true,
83
+ false,
84
+ false,
85
+ false,
86
+ false,
87
+ false,
88
+ true,
89
+ true,
90
+ true,
91
+ false,
92
+ true,
93
+ false,
94
+ false,
95
+ true,
96
+ true,
97
+ false,
98
+ false,
99
+ true,
100
+ true,
101
+ true,
102
+ true,
103
+ true,
104
+ true,
105
+ false,
106
+ false,
107
+ true,
108
+ true,
109
+ false,
110
+ true,
111
+ false,
112
+ true,
113
+ true,
114
+ true,
115
+ true,
116
+ true,
117
+ true,
118
+ false,
119
+ true,
120
+ false,
121
+ true,
122
+ false,
123
+ true,
124
+ true,
125
+ false,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ true,
132
+ false,
133
+ true,
134
+ false,
135
+ true,
136
+ true,
137
+ false,
138
+ true,
139
+ true,
140
+ true,
141
+ true,
142
+ false,
143
+ true,
144
+ true,
145
+ false,
146
+ false,
147
+ true,
148
+ false,
149
+ true,
150
+ true,
151
+ true,
152
+ false,
153
+ true,
154
+ true,
155
+ false,
156
+ true,
157
+ true,
158
+ false,
159
+ true,
160
+ true,
161
+ false,
162
+ true,
163
+ false,
164
+ false,
165
+ true,
166
+ false,
167
+ true,
168
+ false,
169
+ false,
170
+ true,
171
+ true,
172
+ false,
173
+ false,
174
+ false,
175
+ true,
176
+ false,
177
+ true,
178
+ true,
179
+ true,
180
+ true,
181
+ true,
182
+ true,
183
+ false,
184
+ false,
185
+ true,
186
+ true,
187
+ true,
188
+ true,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ true,
195
+ false,
196
+ true,
197
+ true,
198
+ false,
199
+ true,
200
+ true,
201
+ false,
202
+ false,
203
+ false,
204
+ true,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 248.1869854927063,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_75_seed45.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 75,
5
+ "seed": 45,
6
+ "successes": [
7
+ false,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ true,
13
+ true,
14
+ false,
15
+ true,
16
+ false,
17
+ false,
18
+ true,
19
+ false,
20
+ true,
21
+ true,
22
+ false,
23
+ true,
24
+ true,
25
+ true,
26
+ true,
27
+ true,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ true,
33
+ false,
34
+ true,
35
+ false,
36
+ false,
37
+ false,
38
+ false,
39
+ true,
40
+ false,
41
+ false,
42
+ false,
43
+ true,
44
+ false,
45
+ false,
46
+ false,
47
+ true,
48
+ false,
49
+ false,
50
+ false,
51
+ true,
52
+ true,
53
+ true,
54
+ true,
55
+ false,
56
+ true,
57
+ false,
58
+ false,
59
+ false,
60
+ false,
61
+ true,
62
+ true,
63
+ true,
64
+ true,
65
+ false,
66
+ false,
67
+ true,
68
+ false,
69
+ false,
70
+ false,
71
+ false,
72
+ true,
73
+ true,
74
+ false,
75
+ true,
76
+ false,
77
+ true,
78
+ false,
79
+ true,
80
+ true,
81
+ false,
82
+ true,
83
+ false,
84
+ false,
85
+ true,
86
+ true,
87
+ true,
88
+ true,
89
+ false,
90
+ true,
91
+ true,
92
+ true,
93
+ true,
94
+ false,
95
+ false,
96
+ true,
97
+ true,
98
+ true,
99
+ true,
100
+ true,
101
+ true,
102
+ false,
103
+ true,
104
+ true,
105
+ true,
106
+ true,
107
+ false,
108
+ true,
109
+ false,
110
+ false,
111
+ true,
112
+ true,
113
+ false,
114
+ false,
115
+ false,
116
+ true,
117
+ false,
118
+ false,
119
+ false,
120
+ true,
121
+ true,
122
+ false,
123
+ false,
124
+ true,
125
+ false,
126
+ true,
127
+ true,
128
+ true,
129
+ true,
130
+ true,
131
+ true,
132
+ true,
133
+ true,
134
+ false,
135
+ false,
136
+ true,
137
+ true,
138
+ false,
139
+ false,
140
+ true,
141
+ false,
142
+ false,
143
+ false,
144
+ true,
145
+ false,
146
+ true,
147
+ false,
148
+ false,
149
+ true,
150
+ true,
151
+ false,
152
+ true,
153
+ true,
154
+ true,
155
+ true,
156
+ false,
157
+ true,
158
+ false,
159
+ true,
160
+ false,
161
+ false,
162
+ false,
163
+ true,
164
+ true,
165
+ true,
166
+ true,
167
+ false,
168
+ true,
169
+ true,
170
+ true,
171
+ true,
172
+ false,
173
+ false,
174
+ true,
175
+ false,
176
+ true,
177
+ true,
178
+ true,
179
+ false,
180
+ true,
181
+ true,
182
+ true,
183
+ true,
184
+ false,
185
+ true,
186
+ false,
187
+ false,
188
+ false,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ false,
195
+ true,
196
+ true,
197
+ false,
198
+ false,
199
+ false,
200
+ true,
201
+ true,
202
+ true,
203
+ false,
204
+ false,
205
+ true,
206
+ false
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 261.82840752601624,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }
results/pldm_cube_75_seed46.json ADDED
@@ -0,0 +1,312 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "pldm",
3
+ "task": "cube",
4
+ "offset": 75,
5
+ "seed": 46,
6
+ "successes": [
7
+ true,
8
+ true,
9
+ true,
10
+ false,
11
+ true,
12
+ true,
13
+ true,
14
+ true,
15
+ false,
16
+ true,
17
+ false,
18
+ false,
19
+ true,
20
+ false,
21
+ true,
22
+ false,
23
+ false,
24
+ false,
25
+ true,
26
+ false,
27
+ true,
28
+ false,
29
+ true,
30
+ true,
31
+ false,
32
+ false,
33
+ true,
34
+ false,
35
+ false,
36
+ false,
37
+ true,
38
+ true,
39
+ false,
40
+ false,
41
+ true,
42
+ false,
43
+ true,
44
+ true,
45
+ true,
46
+ true,
47
+ false,
48
+ false,
49
+ false,
50
+ false,
51
+ false,
52
+ false,
53
+ true,
54
+ false,
55
+ true,
56
+ true,
57
+ true,
58
+ true,
59
+ false,
60
+ true,
61
+ false,
62
+ false,
63
+ true,
64
+ false,
65
+ false,
66
+ false,
67
+ true,
68
+ false,
69
+ false,
70
+ true,
71
+ false,
72
+ true,
73
+ true,
74
+ false,
75
+ false,
76
+ true,
77
+ false,
78
+ false,
79
+ true,
80
+ false,
81
+ true,
82
+ true,
83
+ true,
84
+ false,
85
+ false,
86
+ true,
87
+ true,
88
+ true,
89
+ true,
90
+ true,
91
+ false,
92
+ false,
93
+ true,
94
+ true,
95
+ false,
96
+ false,
97
+ true,
98
+ true,
99
+ true,
100
+ false,
101
+ false,
102
+ true,
103
+ false,
104
+ true,
105
+ false,
106
+ true,
107
+ true,
108
+ false,
109
+ false,
110
+ true,
111
+ true,
112
+ true,
113
+ false,
114
+ true,
115
+ true,
116
+ true,
117
+ false,
118
+ false,
119
+ false,
120
+ false,
121
+ true,
122
+ false,
123
+ true,
124
+ false,
125
+ false,
126
+ true,
127
+ false,
128
+ false,
129
+ true,
130
+ false,
131
+ true,
132
+ false,
133
+ false,
134
+ true,
135
+ true,
136
+ true,
137
+ false,
138
+ false,
139
+ false,
140
+ false,
141
+ false,
142
+ true,
143
+ false,
144
+ false,
145
+ true,
146
+ true,
147
+ false,
148
+ false,
149
+ false,
150
+ true,
151
+ true,
152
+ false,
153
+ false,
154
+ false,
155
+ true,
156
+ false,
157
+ true,
158
+ true,
159
+ true,
160
+ true,
161
+ true,
162
+ true,
163
+ false,
164
+ true,
165
+ true,
166
+ true,
167
+ true,
168
+ true,
169
+ false,
170
+ false,
171
+ false,
172
+ false,
173
+ true,
174
+ false,
175
+ false,
176
+ true,
177
+ false,
178
+ false,
179
+ false,
180
+ true,
181
+ false,
182
+ true,
183
+ true,
184
+ true,
185
+ true,
186
+ false,
187
+ false,
188
+ false,
189
+ true,
190
+ false,
191
+ false,
192
+ false,
193
+ true,
194
+ true,
195
+ true,
196
+ false,
197
+ false,
198
+ true,
199
+ true,
200
+ true,
201
+ true,
202
+ true,
203
+ true,
204
+ true,
205
+ true,
206
+ true
207
+ ],
208
+ "checkpoint_sha256": "aafc82a80f1ac84928053e153dc67c7d3b1717bcdaa604ce9a3f1deebc842eed",
209
+ "eval_budget": 50,
210
+ "cem": {
211
+ "samples": 300,
212
+ "iterations": 30,
213
+ "topk": 30
214
+ },
215
+ "elapsed_seconds": 258.1439380645752,
216
+ "checkpoint_release": "Locally trained 10 epochs: PLDM stable-worldmodel recipe",
217
+ "training": {
218
+ "method": "pldm",
219
+ "task": "cube",
220
+ "variant": "PLDM stable-worldmodel recipe",
221
+ "epochs_required": 10,
222
+ "training_seed": 42,
223
+ "train_clips": 1638000,
224
+ "total_clips": 1820000,
225
+ "batch_size": 128,
226
+ "batches_per_epoch": 12796,
227
+ "frameskip": 5,
228
+ "history_size": 3,
229
+ "image_size": 224,
230
+ "precision": "bf16-mixed",
231
+ "train_split": 0.9,
232
+ "action_mean": [
233
+ 0.0108846967805138,
234
+ -0.0031414329928013945,
235
+ 0.0026465825646189004,
236
+ 0.0004239286647865428,
237
+ 0.15925256653030173
238
+ ],
239
+ "action_std": [
240
+ 0.2894198325506555,
241
+ 0.39371697495490987,
242
+ 0.6431365217957304,
243
+ 0.3928016202283424,
244
+ 0.2503073640045173
245
+ ],
246
+ "optimizer": "AdamW",
247
+ "lr": 5e-05,
248
+ "weight_decay": 0.001,
249
+ "loss_weights": {
250
+ "std_loss": 18.0,
251
+ "std_t_loss": 0.7,
252
+ "cov_loss": 12.0,
253
+ "cov_t_loss": 0.0,
254
+ "temp_align_loss": 0.2
255
+ },
256
+ "note": "Full epochs of the 90% clip split, no capped training subset. Evaluation pairs remain the shared existing dataset pairs, not a held-out test set.",
257
+ "epochs_completed": 10,
258
+ "history": [
259
+ {
260
+ "epoch": 1,
261
+ "loss": 1.2225470243067844,
262
+ "seconds": 5826.778991699219
263
+ },
264
+ {
265
+ "epoch": 2,
266
+ "loss": 0.6854550508725789,
267
+ "seconds": 25979.50070333481
268
+ },
269
+ {
270
+ "epoch": 3,
271
+ "loss": 0.6379903477219799,
272
+ "seconds": 6363.682020902634
273
+ },
274
+ {
275
+ "epoch": 4,
276
+ "loss": 0.6130627957069341,
277
+ "seconds": 6017.385094881058
278
+ },
279
+ {
280
+ "epoch": 5,
281
+ "loss": 0.5980935352934491,
282
+ "seconds": 6056.753978967667
283
+ },
284
+ {
285
+ "epoch": 6,
286
+ "loss": 0.5877561344275738,
287
+ "seconds": 6000.572838306427
288
+ },
289
+ {
290
+ "epoch": 7,
291
+ "loss": 0.5806439629278171,
292
+ "seconds": 6169.721336841583
293
+ },
294
+ {
295
+ "epoch": 8,
296
+ "loss": 0.5760280382040703,
297
+ "seconds": 5780.607115745544
298
+ },
299
+ {
300
+ "epoch": 9,
301
+ "loss": 0.5730076796116401,
302
+ "seconds": 144145.69498491287
303
+ },
304
+ {
305
+ "epoch": 10,
306
+ "loss": 0.5721370456003703,
307
+ "seconds": 10359.886939525604
308
+ }
309
+ ]
310
+ },
311
+ "smoke": false
312
+ }